Real-time Data Analyzer User's Manual

Real-time Data AnalyzerUser's Manual

-SW1DND-RDA-M

1

SAFETY PRECAUTIONS(Read these precautions before using this product.)

Before using this product, please read this manual and the relevant manuals carefully and pay full attention to safety to handle

the product correctly.

The precautions given in this manual are concerned with this product only. For the safety precautions of the programmable

controller system, refer to the user's manual for the CPU module used.

In this manual, the safety precautions are classified into two levels: " WARNING" and " CAUTION".

Under some circumstances, failure to observe the precautions given under " CAUTION" may lead to serious

consequences.

Observe the precautions of both levels because they are important for personal and system safety.

Make sure that the end users read this manual and then keep the manual in a safe place for future reference.

[Design Precautions]

[Design Precautions]

WARNING● To perform an operation, such as data change or operating status change, to running devices, such as

a programmable controller, servo, robot, or server, from an industrial PC equipped with this product,

configure an interlock circuit outside of the devices so that the entire system always operates to the

safety side. Additionally, read this manual carefully and ensure the safety before operations.

Especially, in the above mentioned operations that are performed from external devices through

network, any problems on devices may not be dealt with promptly due to an abnormal data

communication.

● Configure a safety circuit outside of an industrial PC equipped with this product so that the entire

system operates to the safely side even when a fault occurs in the PC.

Failure to do so may result in an accident due to an incorrect output or malfunction.

CAUTION● During application of each setting, do not perform the operation that forcibly turns the industrial PC

equipped with this product OFF.

Otherwise, the data will be undefined and resetting and re-registering data will be required.

Additionally, doing so may cause the malfunction of this product.

WARNING Indicates that incorrect handling may cause hazardous conditions, resulting in death or severe injury.

CAUTION Indicates that incorrect handling may cause hazardous conditions, resulting in minor or moderate injury or property damage.

2

[Security Precautions]

WARNING● To maintain the security (confidentiality, integrity, and availability) of access target devices and the

system against unauthorized access, denial-of-service (DoS) attacks, computer viruses, and other

cyberattacks from external devices via the network, take appropriate measures such as firewalls,

virtual private networks (VPNs), and antivirus solutions.

3

CONDITIONS OF USE FOR THE PRODUCT

(1) This software shall be used under the following conditions;i) that any failure occurred in this software, if any, shall not lead to any serious accident.ii) that the backup and/or fail-safe functions are systematically performed outside the devices in the cases of any failure occurred in this software.

(2) Mitsubishi Electric assumes no responsibility and liability (including but not limited to, default liability, defect liability, quality assurance responsibility, tort liability, product liability) for the quality, performance, and safety of both this software and products using this software.

(3) Mitsubishi Electric shall have no responsibility or liability for any problems involving this software and system trouble caused by DoS attacks, unauthorized access, computer viruses, and other cyberattacks.

4

INTRODUCTIONThank you for purchasing Real-time Data Analyzer.

This manual describes the performance specifications, procedures before operation, and troubleshooting of this product.

Before using this product, please read this manual and the relevant manuals carefully, and develop familiarity with the

functions and performance of this product to handle correctly.

Please make sure that the end users read this manual.

5

CO

NT

EN

TS

CONTENTSSAFETY PRECAUTIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1

CONDITIONS OF USE FOR THE PRODUCT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3

INTRODUCTION. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4

RELEVANT MANUALS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .9

TERMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .9

PART 1 Real-time Data Analyzer

CHAPTER 1 Real-time Data Analyzer 12

CHAPTER 2 SPECIFICATIONS 14

2.1 Functional Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

Similar Waveform Recognition Tool. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

Real-time Statistic Diagnosis Tool . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

2.2 Operational Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

Operating status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

Operation when an error occurs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

CHAPTER 3 FUNCTION LIST 20

3.1 Function List of Similar Waveform Recognition Tool . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

3.2 Function List of Real-time Statistic Diagnosis Tool. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

CHAPTER 4 PROCEDURE BEFORE OPERATION 23

4.1 License Manager . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

Startup method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

Screen configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

Registration procedure of a temporary license . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28

Registration procedure of a subscription license. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28

4.2 Startup Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

Similar Waveform Recognition Tool. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

Real-time Statistic Diagnosis Tool . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

4.3 Display Language Switching . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

PART 2 Similar Waveform Recognition Tool

CHAPTER 5 OVERVIEW 32

5.1 Home screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

Setting file import. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34

Setting file export. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34

Edgecross Link Setting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35

CHAPTER 6 WAVEFORM LEARNING 36

6.1 Operation Overview in Waveform Learning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

CHAPTER 7 DATA DIAGNOSIS 38

7.1 Operation Overview in Similarity Diagnosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40

6

7.2 Similarity Score . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41

7.3 Similarity Score Threshold List . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

7.4 Pause of Similarity Diagnosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

CHAPTER 8 PROCEDURE TO USE Similar Waveform Recognition Tool 44

8.1 Data Analysis Setting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45

8.2 Data Diagnosis Setting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50

8.3 Starting a Temporary Operation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54

8.4 Starting an Actual Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57

Checking the diagnosis status after starting an actual operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57

Changing any settings after starting an actual operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58

CHAPTER 9 CREATING REFERENCE WAVEFORM LEARNING DATA 62

9.1 Reference Waveform Learning Data File Creation Setting Screen. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63

CHAPTER 10 PUBLISHING DATA SETTING 65

10.1 Publishing Data Setting Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66

10.2 Publishing Data List Screen. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67

CHAPTER 11 OPERATION SETTING 68

11.1 Operation Setting Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69

11.2 Publishing Data List Screen. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73

11.3 Diagnosis Stop Condition Setting Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74

CHAPTER 12 RESPONSE DATA SETTING 75

12.1 Response Data List Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76

CHAPTER 13 EXECUTION/MONITORING OF SIMILAR WAVEFORM RECOGNITION 77

13.1 Similar Waveform Recognition Execution Status Detailed Monitoring Screen. . . . . . . . . . . . . . . . . . . . . . 79

CHAPTER 14 OUTPUTTING DIAGNOSIS RESULT FILE 81

14.1 File Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82

14.2 File Structure (When "Only difference condition" is selected for "Output Settings") . . . . . . . . . . . . . . . . 83

14.3 File Structure (When "Fixed output cycle" is selected for "Output Settings") . . . . . . . . . . . . . . . . . . . . . . 85

14.4 Operational Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87

CHAPTER 15 GX LogViewer COOPERATION 90

15.1 Similar waveform recognition monitor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90

15.2 Diagnosis result file display . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91

CHAPTER 16 CONSIDERATIONS FOR SIMILAR WAVEFORM RECOGNITION 92

16.1 Waveforms Used for Similar Waveform Recognition. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92

16.2 Input Waveform (Reference Waveform) and Setting Items in Waveform Learning. . . . . . . . . . . . . . . . . . . 94

16.3 Setting Items for a Data Diagnostic Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94

16.4 Restrictions on Reference Waveform Learning Data Once Created . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95

16.5 Number of Times of Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95

16.6 Relativity between Collection Data Transmission Delay and a Diagnosis Result. . . . . . . . . . . . . . . . . . . . 96

CHAPTER 17 TROUBLESHOOTING BY SYMPTOM 98

7

CO

NT

EN

TS

PART 3 Real-time Statistic Diagnosis Tool

CHAPTER 18 OVERVIEW 108

18.1 Configuration Software. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109

Screen configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110

CHAPTER 19 PROCEDURE TO USE Real-time Statistic Diagnosis Tool 114

19.1 Data Analysis Setting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114

19.2 Data Diagnosis Setting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117

19.3 Migrating settings to Real-time Statistic Diagnosis Tool in another industrial PC. . . . . . . . . . . . . . . . . . 121

Procedure to be performed in a migration source industrial PC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121

Procedure to be performed in a migration destination industrial PC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121

19.4 Project File Setting Utility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122

CHAPTER 20 Reader 125

20.1 Screen Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126

20.2 Right-click Menu Items . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128

20.3 Folder Definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129

20.4 Adding a Unit Definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131

20.5 Editing Settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142

20.6 Switching the Operation Mode to CSV File Reading Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162

CHAPTER 21 Analyzer 164

21.1 Displaying Read Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165

21.2 Displaying Read Data Statistics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180

21.3 Displaying Read Data Frequency. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183

21.4 SPC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189

21.5 Multivariate Analysis (Single Central Value Format) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207

21.6 Guard Band Diagnosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215

21.7 Creating Correlation Matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222

21.8 Multiple Regression Analysis (LMR) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243

21.9 Maharanobis - Taguchi method (MT) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 250

21.10 Editing Logics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255

21.11 Setting logic variable . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 356

21.12 Displaying Data after Running . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 365

21.13 Displaying a Result of Simple diagnosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 374

21.14 Displaying a Diagnostic Logic Result . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 382

21.15 Management. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 384

21.16 Data Management Setting. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 390

21.17 Waveform Display Setting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 391

21.18 Logic Setting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 393

21.19 Programmable Controller CPU Cooperation Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 394

Guard band diagnosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 394

Multiple regression analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 402

Maharanobis-Taguchi method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 409

CHAPTER 22 Monitor 416

22.1 Screen Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 417

22.2 Monitor Setting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 419

8

22.3 Starting Monitor. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 421

CHAPTER 23 TROUBLESHOOTING BY SYMPTOM 422

APPENDIX 424

Appendix 1 Processing Performance and Processing Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 424

Similar Waveform Recognition Tool. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 424

Real-time Statistic Diagnosis Tool . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 432

Appendix 2 Available Characters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 433



Appendix 3 Environment Settings for an Industrial PC where Real-time Data Analyzer Runs . . . . . . . . . . . . . 440

Settings for indexes created for Windows Search. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 440

Settings for the Customer Experience Improvement Program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 443

Appendix 4 System Configuration and Application Settings for Operation of Real-time Data Analyzer . . . . . 444

System configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 444

Application settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 445

Corrective actions for high load on an industrial PC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 446

Appendix 5 Software Licenses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 447

Software information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 447

Appendix 6 Supported Edgecross Basic Software Version. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 457



Appendix 7 Added and Changed Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 458

INDEX 459

REVISIONS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .461

TRADEMARKS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .462

9

RELEVANT MANUALS

e-Manual refers to the Mitsubishi Electric FA electronic book manuals that can be browsed using a dedicated

tool.

e-Manual has the following features:

• Required information can be cross-searched in multiple manuals.

• Other manuals can be accessed from the links in the manual.

• Hardware specifications of each part can be found from the product figures.

• Pages that users often browse can be bookmarked.

• Sample programs can be copied to an engineering tool.

TERMSUnless otherwise specified, this manual uses the following terms.

Manual name [manual number] Description Available form

Real-time Data Analyzer User's Manual

[SH-081873ENG] (this manual)

Specifications, procedure before operation, functions, and troubleshooting of

Real-time Data Analyzer

e-Manual

PDF

Term Description

Business system A system that supports business management such as sales management and production management.

Correlation matrix A matrix created based on the reference CSV file. It is used for advanced analysis.

Data collection Processing to collect data for data diagnosis and data analysis.

Data collector A software component that collects data in production sites through each network. It is provided by vendors for each

network and connection target devices.

Data diagnosis Processing to diagnose if collected data matches preset conditions.

Data modification Processing to modify collected data into a form suitable for analysis and diagnostic.

Further modifications can also be made to the data modified once.

Dataset A group of data in a read CSV file. It is used for a logic.

Detection It is to indicate that waveforms are different, as a diagnosis result.

Detection sensitivity The difficulty of determination that a waveform is similar or dissimilar, as a result of similarity diagnosis.

It is difficult to make the determination as the detection sensitivity is higher. (The proportion of waveforms determined to

be different increases.)

Diagnosis result A diagnosis result data. (0: similar, 1:different)

Diagnosis result data A general term for information on a diagnostic result passed to Real-time Flow Manger after the similarity diagnosis.

Different state A diagnosis result that indicates 'similarity score < similarity score threshold'.

Edge application Software that performs various processing for data utilization in production sites using the functions provided by

Edgecross in the edge computing area.

Edge computing An information processing method and area, for not only collecting and analyzing data in production sites in real-time,

and feed backing the data, but summarizing the production site data and sharing information with IT systems efficiently

with the hardware and software in production sites.

Edgecross A software platform that implements specifications and concepts for realizing manufacturing solutions by the FA-IT

collaboration centering on the edge computing.

Edgecross Basic Software The name of the software product that implements the Edgecross function.

Feedback Processing to report a detection result when diagnosed data matches a condition.

GX LogViewer A software product that has a function to display an inspection waveform and the similarity scores on a graph in real time

and a function to display the graph of a diagnosis result file.

Historical data definition file A CSV file that stores the definition information of a reference waveform file.

Industrial PC A PC specialized for industrial use with features such as high reliability, environmental resistance, and long-term supply.

Inspection waveform Waveform data to be diagnosed.

Internal database Data group that software included in Real-time Statistic Diagnosis Tool uses to share data.

MQTT An abbreviation for MQ Telemetry Transport.

MQ Telemetry Transport is a lightweight message communication protocol using TCP/IP, and the standard is published

as an open protocol.

MQTT broker An application that acts as an intermediary between applications that distribute messages and ones that receive the

messages by using MQTT.

10

Partial matrix A matrix that includes only necessary data extracted by removing incorrect variables and exclusive variables from the

whole matrix.

Process A generic term for the processing of data collection, data modification, data diagnosis, and feedback that compose a

process flow.

Process flow An execution unit of sequential processing performed by Real-time Flow Manager, consisting of data collection, data

modification, data diagnosis, and feedback.

Processing flow execution control A control processing that executes and controls a series of processing flow performed in the process for realizing a

certain purpose or application using data, and processes the data efficiently (the data includes stream data).

Publishing data definition file A file that stores the information to set the publishing data to an Edge application in advance.

Real-time data processing A function that realizes the real-time diagnosis and feedback of the production site data.

It is composed of the processing flow execution control and stream data processing.

Real-time Flow Designer The name of the software component that performs operation setting of Real-time Flow Manager.

Real-time Flow Manager The name of the Windows version software component that implements the real-time data processing.

Reference waveform A group of one or more reference waveform files in which waveforms to be referenced for diagnosis are recorded.

The main purpose of a waveform similarity diagnosis is determining whether a waveform is similar with a reference

waveform or not.

Reference waveform file A CSV file in which waveforms to be referenced for diagnostic is recorded.

Reference waveform learning data Data that stores the learning result of a reference waveform.

The property of the reference waveform is recorded. By using this data for the similarity diagnosis instead of a reference

waveform, the similarity equivalent to a comparison using the reference waveform can be determined at high speed.

Reference waveform learning data

file

A file which stores reference waveform learning data.

Response data definition file A file that stores the information in order to send response definitions to Real-time Flow Designer.

Response data file A CSV file which is output when an edge application completed a data analysis and detected an error.

Section The unit of data used in Real-time Statistic Diagnosis Tool.

In Real-time Statistic Diagnosis Tool, an analysis or diagnosis is performed for each section.

Setting file A file that saves the setting contents edited in Similar Waveform Recognition Tool (extension: .swrcfg).

Similar state A diagnosis result that indicates 'similarity score similarity score threshold'.

Similar waveform recognition Refers to the following functions:

• Learning a reference waveform and creating a reference waveform learning data

• Interface to enter an inspection waveform or output diagnosis result data

• Diagnosing the similarity between a reference waveform and an inspection waveform at high speed by using

reference waveform learning data

• Outputting a diagnosis result in a CSV file

Similarity diagnosis Processing to obtain the similarity between a reference waveform and an inspection waveform at high speed by using

information of the reference waveform recorded in reference waveform learning data.

Stream data processing A function that cuts data continuously generated in the appropriate unit for data analysis and distributes it to the edge

application.

Unit A unit of a setting in Real-time Statistic Diagnosis Tool.

It can be defined and used for each CSV file format to be read.sd

Waveform learning Processing to learn the property of a reference waveform and create reference waveform learning data in which the

learning information is recorded.

Whole matrix The original matrix of a created correlation matrix.

Term Description

11

PA

RT

1

PART 1 Real-time Data Analyzer

This part explains Real-time Data Analyzer.

Page 12 Real-time Data Analyzer

Page 14 SPECIFICATIONS

Page 20 FUNCTION LIST

Page 23 PROCEDURE BEFORE OPERATION

121 Real-time Data Analyzer

1 Real-time Data Analyzer

Real-time Data Analyzer is an edge application that analyzes the data of a production site offline and diagnose the data in real

time.

Real-time Data Analyzer consists of the following edge applications:

By installing this product in an industrial PC and using with Real-time Flow Manager, the data collected by Real-time Flow

Manager is analyzed offline and diagnosed in real time. (Page 12 Offline analysis, Page 13 Real-time diagnosis)

For Real-time Flow Manager, refer to the following manual.

Edgecross Basic Software for Windows User's Manual

Offline analysisThis creates rules to enable users to analyze and diagnose data collected from a production site.

The following figure describes the flow of offline analysis.

Real-time Flow Manage outputs the data collected in the data collection process as a file. (Data will be modified in the data

modification process as necessary.)

A user creates diagnosis rules in Real-time Data Analyzer based on the file of the output collection data.

Edge application Description

Similar Waveform Recognition Tool AI-equipped edge application. This application calculates the similarity between the waveform to be referenced and an

inspection waveform at high-speed and detects the differences of the inspection waveform by using the data created

by learning the reference waveform.

Real-time Statistic Diagnosis Tool Edge application that diagnoses the waveform data according to the rules which is derived in a statistic method or

multivariate analysis and determines whether the data is OK or NG.

(1) Real-time Data Analyzer

(2) Diagnosis rule

(3) Real-time Flow Manager

(1)

(3)

� �

�

(2)

Data collection

Data modification

Industrial personal computer

1 Real-time Data Analyzer 13

1Real-time diagnosisThis diagnoses data collected from a production site in real time according to the diagnosis rules created in offline analysis.

The following figure describes the flow of real-time diagnosis.

Real-time Flow Manage outputs the data collected in the data collection process. (Data will be modified in the data

modification process as necessary.)

Real-time Data Analyzer diagnoses collected data which was output by Real-time Flow Manager according to diagnosis

rules created in offline analysis.

Real-time Data Analyzer outputs the diagnosis result data.

Real-time Flow Manager sends a feedback based on the diagnosis result data.


(2) Diagnosis rule


(3)

��

�

(1)

�

�

(2)

Industrial personal computer

Data modification

Data collection

Feedback execution

142 SPECIFICATIONS2.1 Functional Specifications

2 SPECIFICATIONS

This chapter shows the functional and operational specifications of Real-time Data Analyzer.

2.1 Functional SpecificationsThis section shows the functional specifications of Real-time Data Analyzer.

Similar Waveform Recognition ToolThe following shows the functional specifications of Similar Waveform Recognition Tool.

Item Specification

Waveform

learning function

Reference

waveform file

File format CSV file

Number of files*1 1 or more (no limitation)

Reference

waveform

Lower limit of number of records*2,*3 • To satisfy the restrictions of the waveform learning function: Records

should be more than 4 times of the unit waveform width and temporally

consecutive in one or more sections.

• To enable the minimum diagnosis operation: Records in which the

basic cycle of a reference waveform is 4 cycles or more and temporally

consecutive in one or more sections.

• To enable the practical operation: Records in which the basic cycle of a

reference waveform is 16 cycles or more and temporally consecutive in

one or more sections.

Upper limit of number of records 1000000

Data type*4 Numerical type (decimal notation or exponent notation)

Value range*5 • To satisfy the restrictions of the waveform learning function: -

2.11996057443428e+152 to 2.11996057443428e+152

• Practical value range: -9.00719925474099e+12 to

9.00719925474099e+12

2 SPECIFICATIONS2.1 Functional Specifications 15

2

Data diagnosis

function

Input data

(Publishing data)

Inspection

waveform

Description Waveform data which is sent from Real-time Flow Manager and to be

diagnosed in a similarity diagnosis

Data name Any name (Set in Real-time Flow Designer)

Data type • INT

• UINT

• DINT

• UDINT

• REAL

• LREAL

Number of input 1 to 64

Output data*6

(Response data)

Number of

different detection

(number of

diagnosis results

determined to be

different)*7

Description Number of diagnosis results determined to be different among all

similarity diagnoses in progress

It will be '0' when all diagnosis results are determined to be similar. In

other cases, it will be '1' or more according to the number of diagnosis

results determined to be different.*8

Data name DifferenceCount

Data type UINT

Number of output 1

Value range • Maximum value: Number of operation settings applied to a similarity

diagnosis

• Minimum value: 0

Relative similarity

score (gap of a

similarity score

and a similarity

score threshold)*9

Description Gap between a similarity score and similarity score threshold acquired in

each individual similarity diagnosis

The value will be '0' or more in a similarity diagnosis determined to be

similar, or lower than '0' in a similarity diagnosis determined to be

different.

Data name Any name (Set in Similar Waveform Recognition Tool)

Data type REAL

Number of output 1 to 64 (Same as the input number of an inspection waveform)

Value range • Maximum value: 100 - similarity score threshold

• Minimum value: 0 - similarity score threshold

Diagnosis result Description Individual similarity diagnosis result

Data name Any name (Set in Similar Waveform Recognition Tool)

Data type UINT

Number of output 1 to 64 (Same as the input number of an inspection waveform)

Value range • 0: Similar

• 1: Different

Item Specification


*1 Sequentially numbered files are read in the file name order. When records stored in the successively read files are continuous in time, it is considered as one unit of waveform data.

*2 If the temporally consecutive record is less than 4 times of the unit waveform width, the section can not be analyzed and is discarded.*3 When the stability of a waveform is low, the diagnosis precision is decreased and a false detection may be reported because the

tendency of waveforms can not be grasped within the number of records mentioned in this item. To solve this issue, increase the number of records used for the learning.

*4 The data types that there are 500 steps or more from the minimum value to the maximum value are recommended.*5 When a waveform that contains an extremely large or small value is diagnosed, a small change may not be applied to the similarity

score, or the similarity score may greatly be reduced because a slight change is considered as a large one.*6 It is used as input data for the feedback function of Real-time Flow Manager.*7 To provide feedback when at least one difference is detected in a result of similarity diagnosis, set the condition to 'DifferenceCount > 0'.*8 When the diagnosis status is in the 'preparing for diagnosis' or 'suspended' state, the number of diagnosis results determined to be

different is counted as shown below according to each situation.Before a diagnosis starts: The number is not counted since the diagnosis is not yet performed.When a diagnosis status changed from 'similar' to 'suspended' or 'preparing for diagnosis': The number is not counted since the diagnosis status is determined to be similar.When a diagnosis status changed from 'different' to 'suspended' or 'preparing for diagnosis': The number is counted since the diagnosis status is determined to be different.

*9 To provide feedback when a difference is detected in a result of specific similarity diagnosis, set the condition to 'Gap of a similarity score and a similarity score threshold < 0'.

Real-time Statistic Diagnosis ToolThe following shows the functional specifications of Real-time Statistic Diagnosis Tool.

In the following table, MVA refers to the multivariate analysis (single central value format) of the simple analysis and simple

diagnosis, and GB refers to the guard band.

Item Specification

Response data

file

File name (Unit name)_(YYYY_mm_dd HH_MM_ss.ms*1).csv

Output data TIME Data type DATETIME

Output content End time of a diagnosis section (YYYY/MM/DD hh:mm:ss.ms)

RESULT Data type BOOL

Output content Value (1) indicating that an error is detected in a diagnosis

TYPE Data type INT

Output content Numeric value indicating the diagnosis type

• 1: Simple diagnosis (SPC)

• 2: Simple diagnosis (Guard band)

• 3: Simple diagnosis (Multivariate analysis (single central value

format), calculation type = T2)


format), calculation type = Q)


format), calculation type = KNN)


format), calculation type = MT)


format), calculation type = EU)

• 10: Analysis logic, diagnostic logic

NAME Data type STRING

Output content • Simple diagnosis:

Diagnosis name specified in a simple diagnosis*2

• Analysis logic, diagnostic logic:

Logic name*2

2 SPECIFICATIONS2.1 Functional Specifications 17

2

Diagnosis result

detail information

file

File name (Unit name)_(diagnosis type (SPC, GB, or MVA))_(time at which an

error was detected (YYYY_mm_dd HH_MM_ss_ms*1)).csv*3

File format Delimiter Comma (,)

Line feed code CRLF(0x0D, 0x0A)

Character code Shift JIS

Number of columns 13 to 51 columns

Number of rows 2 rows (header row and data row)

Output data EQ*4 Data type STRING

Output content*5 Equipment

UNIT*4 Data type STRING

Output content*5 Unit name

SECTION*4 Data type STRING

Output content*5 Extended section name (blank if not set)

DIAG_NAME*4 Data type STRING

Output content*5 Diagnosis name

DIAG_TYPE*4 Data type STRING

Output content*5 Diagnosis type (SPC, GB, or MVA)

STATISTICS*4 Data type STRING

Output content*5 • SPC and MVA: Type of selected statistics

• GB: Blank

TYPE*4 Data type STRING

Output content*5 SPC: Decision rule (3 Default SPC rule)

MVA: Calculation type*6*7*8 (Page 208 Screen configuration)

GB: Condition of fault judgement sections (It is output in the format

of "(number of error occurrences)/(number of judgement

sections).")

THRESHOLD*4

(For GB,

'THRESHOLD**' ('**' is

a two-digit value

starting with '00.'))

Data type LREAL

Output content*5 SPC: Threshold values (1 to 5 sigma, management upper and lower

limits, or toward center) when a decision rule is set to use the

threshold values for judgement*9

MVA: Upper limit set as a reference

GB: Guard band value at the time when an error was detected (A

guard band value corresponding to the value in the "VALUE**"

column is output for each error section. In addition '**' in the column

name "THRESHOLD**" is the value same as '**' in the

corresponding "VALUE**" column.)

START_TIME*4 Data type DATETIME

Output content*5 Section start time (YYYY/mm/dd HH:MM:ss.ms*1)

END_TIME*4 Data type DATETIME

Output content*5 Section end time (YYYY/mm/dd HH:MM:ss.ms*1)

DETECT_TIME*4 Data type DATETIME

Output content*5 • SPC and MVA: Blank (Data time is displayed with the section

start and end time since it is determined by calculating the

statistics of the section.)

• GB: Time at which an error was detected (YYYY/mm/dd

HH:MM:ss.ms*1)

VARIABLE*4 Data type STRING

Output content*5 Diagnosis target variable name (It is output by delimiting the

variable names of calculation sources with '/' if the value of a

diagnosis target is calculated from multiple variables.)

Item Specification


*1 Up to three digits are output.*2 To cooperate with Real-time Flow Manager, set the name from 1 to 32 characters.*3 (Example) When a unit name is 'S001': S001_SPC_2017_03_15_19_47_53_789.csv*4 Output in the header row.*5 Output in the data row.*6 When the calculation type is T2 or Q, the calculation type is output followed by the number of principal components which is output in the

format of "/(number of principal components)."*7 The calculation type (the number of principal components if the calculation type is T2 or Q) is output followed by the condition of fault

judgement section which is output in the format of "(number of error occurrences)/(number of judgement sections)."*8 (Example) T2: T2/2(2/3)

(Example) MT: MT(2/3)*9 Blank if a decision rule is set not to use the threshold values for judgement.

Diagnosis result

detail information

file

Output data VALUE***4

'**' is a two-digit value

starting with '00.'

Data type LREAL

Output content*5 Value of a diagnosis target

■Details on the value

SPC: The output value differs depending on the decision rule.

• Other than 1 to 5 sigma: Sigma value of statistics calculated for

each section

• Other rules: Statistics calculated for each section

MVA: Result of calculation selected for the calculation type

GB: Value measured when an error was detected

■Number of output columns

When a diagnosis requires multiple fault judgement sections, the

value is output for the number of the following sections by setting

each column name as "VALUE**." (Up to 20 sections (VALUE00 to

VALUE19))

• SPC and MVA: Fault judgement sections

• GB: Sections in which an error occurred

Item Specification

2 SPECIFICATIONS2.2 Operational Specifications 19

2

2.2 Operational SpecificationsThis section shows the operational specifications of Real-time Data Analyzer.

Operating statusThe following shows the operating status of Real-time Data Analyzer.

Similar Waveform Recognition ToolThe following table shows the operating status of Similar Waveform Recognition Tool.

Functions that can be executed vary depending on the operating status of Similar Waveform Recognition Tool.

: Executable, : Not executable

Real-time Statistic Diagnosis ToolThe following table shows the operating status of Real-time Statistic Diagnosis Tool.

In Real-time Statistic Diagnosis Tool, the software configuring the real time statistical diagnostics tool, Reader, Analyzer, and

Monitor, has the operating status individually.

For the software configuration, refer to the following:

Page 109 Configuration Software

■ Reader

■ Analyzer

■ Monitor

Operation when an error occursWhen an error occurs in Real-time Data Analyzer, an error message is displayed on the top of the screen.

Status Description Function

Waveform learning

Similarity diagnosis setting

Diagnosis status display

Connection with GX LogViewer

Stop A state in which the similarity diagnosis is

in stand-by

Run A state in which the similarity diagnosis is

in operation

Status Description

Setting editable mode A state in which the settings of Reader can be edited

CSV file reading mode A state waiting for storing a CSV file in a specified folder

A CSV file will be read automatically after stored in a specified folder

Status Description

Stopped A state in which either a diagnosis rule is not set or a diagnosis is not performed even after a CSV file

was read by Reader

Diagnosing A state in which the CSV file is read by Reader and a diagnosis can be performed according to the set

diagnosis rule

Status Description

Stopped A state in which the response data file is not output even when an error is detected in a diagnosis

Diagnosing A state in which the response data file when an error is detected in a diagnosis

203 FUNCTION LIST3.1 Function List of Similar Waveform Recognition Tool

3 FUNCTION LIST

This section shows the function list of Real-time Data Analyzer.

3.1 Function List of Similar Waveform Recognition Tool

This following table shows the function list of Similar Waveform Recognition Tool.

An offline analysis and real-time diagnosis are included in the waveform learning function and the data diagnosis function

respectively.

Function Description Reference

Waveform learning function To create the reference waveform learning data used for similar waveform

recognition.

It extracts unit waveforms from a reference waveform for diagnosing the

similarity.

Page 36 WAVEFORM LEARNING

Data diagnosis function To monitor an inspection waveform and notify Real-time Flow Manager of a

waveform determined to be different, that is the similarity score is lower

than a threshold value, if found.

In addition, monitoring of an inspection waveform can be paused by setting

a condition for its publishing data that was input from Real-time Flow

Manager.

Page 38 DATA DIAGNOSIS

GX LogViewer interaction function To display the diagnostic status of a similar waveform recognition in GX

LogViewer.

Page 90 GX LogViewer COOPERATION

3 FUNCTION LIST3.2 Function List of Real-time Statistic Diagnosis Tool 21

3

3.2 Function List of Real-time Statistic Diagnosis ToolThis following table shows the function list of Real-time Statistic Diagnosis Tool.

Offline analysis and real-time diagnostic is included in the data analysis/diagnostic function.


CSV file reading function To read a CSV file output to a specified folder. Page 125 Reader

Data analysis/

diagnostic function

Display function To display the data of a read CSV file.

Read data display To display data of a CSV file as a waveform.

Analysis by overlapping or connecting data can be

performed.

Page 165 Displaying

Read Data

Read data statistics

display

To calculate and display the statistics of CSV file data.

The changes of statistics and correlation of data can be

analyzed.

Page 180 Displaying

Read Data Statistics

Read data frequency

display

To perform STFT conversion and display the spectrogram,

or perform Wavelet conversion and display the scalogram

on data in a CSV file.

Facility failure can be analyzed by visualizing the

frequencies.

Page 183 Displaying

Read Data

Frequency

Simple analysis/

diagnosis function

To analyze and diagnose data by operating GUI.

SPC To calculate the statistics for each unit of collected data and

diagnose it according to the SPC rule.

A sign of failure can be detected based on the changes of

statistics.

Page 189 SPC

Multivariate analysis

(single central value

format)

To detect an event to be a trigger based on the model of the

multivariate analysis result.

Page 207

Multivariate Analysis

(Single Central

Value Format)

Guard band diagnosis To diagnose data by using a guard band created based on

the normal waveform data.

The upper and lower limit values can be specified without

using the reference waveform.

Page 215 Guard

Band Diagnosis

Advanced analysis

function

To perform multivariate analysis by the correlation analysis,

multiple regression analysis, or Mahalanobis-Taguchi

method.

Correlation matrix

creating

To create a correlation matrix for correlation analysis. Page 222 Creating

Correlation Matrix

Multiple regression

analysis (LMR)

To perform multiple regression calculation by selecting one

objective variable and multiple explanatory variables to

obtain the correlation between multiple variables.

Page 243 Multiple

Regression Analysis

(LMR)

Maharanobis -

Taguchi method (MT)

To collect the sample of multiple variables to be referenced

and calculate the Mahalanobis distance for the standard

section and the changes in variation based on the

correlation of multiple variables.

Page 250

Maharanobis -

Taguchi method

(MT)

Analysis logic -

diagnostic logic

operation function

Functions to create arbitrary analysis logic/diagnostic logic

Logic editing To edit the analysis and diagnostic logics. Page 255 Editing

Logics

Logic variable setting To edit variables used in the analysis and diagnostic logics. Page 356 Setting

logic variable

Data display function after execution To display multiple waveforms in a single area.

Analysis with the display method by connecting multiple

sections of single data or overlapping different data in the

same section can be performed.

Page 365 Displaying

Data after Running

223 FUNCTION LIST3.2 Function List of Real-time Statistic Diagnosis Tool


Data analysis/

diagnostic function

Diagnostic result

display function

To display the simple diagnosis result or diagnostic logic

execution result.

Simple diagnosis

result display

To display the result of simple diagnostic (SPC, multivariate

analysis, or guard band diagnosis).

Page 374 Displaying

a Result of Simple

diagnosis

Diagnostic logic result

display

To display the result of diagnostic logic (expansion trace

GB, SPC diagnostic, log writing, and error notification).

Page 382 Displaying

a Diagnostic Logic

Result

Management function To manage data used in Real-time Statistic Diagnosis Tool.

Data to be displayed can be selected or narrowed down.

Page 384

Management

Option setting function To set the setting on Real-time Statistic Diagnosis Tool.

Data management

setting

To set the setting on data management. Page 390 Data

Management

Setting

Waveform display

setting

To set the setting on the waveform display. Page 391 Waveform

Display Setting

Logic setting To set the setting on logics. Page 393 Logic

Setting

Programmable controller cooperation function To diagnose data in a programmable controller in real time

according to the diagnosis rule created in Real-time

Statistic Diagnosis Tool.

Page 394

Programmable

Controller CPU

Cooperation

Function

Diagnostic result notification function To issue an alarm when an error is detected in a diagnostic

result.

Page 416 Monitor

4 PROCEDURE BEFORE OPERATION 23

4

4 PROCEDURE BEFORE OPERATION

This chapter shows the procedure from start-up to operation of Real-time Data Analyzer.

Operating procedure

1. Install Real-time Data Analyzer in an industrial PC.

For the installation procedure and the operating environment of Real-time Data Analyzer, refer to the following document.

Real-time Data Analyzer Installation Instructions

2. Start License Manager and register the license of the Real-time Data Analyzer.

Page 24 License Manager

3. Start Similar Waveform Recognition Tool or the Real-time Statistic Diagnosis Tool.

Page 29 Startup Method

To switch the display language of Similar Waveform Recognition Tool or Real-time Statistic Diagnosis Tool, use the utility for

switching language.

For details, refer to the following:

Page 29 Display Language Switching

4. Install Edgecross Basic Software in the industrial PC.

It can also be installed in an industrial PC in which Real-time Data Analyzer is not installed.

5. Start the operation of Similar Waveform Recognition Tool or Real-time Statistic Diagnosis Tool.

Page 44 PROCEDURE TO USE Similar Waveform Recognition Tool, Page 114 PROCEDURE TO USE Real-time

Statistic Diagnosis Tool

Exclusion setting in antivirus softwareWhen running Real-time Data Analyzer on an industrial PC where an antivirus software is installed, the antivirus software may

affect the Real-time Data Analyzer behavior. Exclude the following folders from the monitoring targets in the antivirus software

settings.

For the setting methods, refer to the manual for each antivirus software.

• Installation destination folder for Real-time Data Analyzer

• Storage destination folder for a reference waveform file*1(Page 63 Reference Waveform Learning Data File Creation

Setting Screen)

*1 Exclude this when learning a waveform.

• Output destination folder for a reference waveform learning data file(Page 63 Reference Waveform Learning Data File

Creation Setting Screen)

• Save destination folder for a diagnosis result file*2(Page 69 Operation Setting Screen)

*2 Exclude this when diagnosing similarity.

• Following files and folders of Windows

Pagefile.sys

*.pst

%systemroot%\System32\Spool

%systemroot%\SoftwareDistribution\Datastore

%windir%\Security\Database

%allusersprofile%\NTUser.pol

%systemroot%\system32\GroupPolicy\Machine\registry.pol

%systemroot%\system32\GroupPolicy\User\registry.pol

• Reader equipment data folder (Page 129 Folder Definition)

• CSV input folder (Page 129 Folder Definition)

• RSD dataset folder (Page 129 Folder Definition)

For the folders which should be excluded from the monitoring targets among the folders which are set by using Edgecross

Basic Software, refer to the following manual.


244 PROCEDURE BEFORE OPERATION4.1 License Manager

4.1 License ManagerLicense Manager is used to register and delete the license of Real-time Data Analyzer and check the status.

The license of Real-time Data Analyzer is authenticated in the subscription method.

The type of license is as follows:

Precautions

• Make sure that the clock setting on Windows is correct before registering a license file for the first time to an industrial PC

in which Real-time Data Analyzer is installed. If the set time on Windows is shifted forward or backward for 24 hours or

more after a license file is registered, the license authentication will fail.

• The license expires at 0:00:00 on the next day of its expiration date. The license expiration is checked by using UTC time

as a standard. Therefore, it is not affected by daylight saving time. (For a temporary license, it will expire on the next day of

the date when the license expiration date becomes less than one day (at 0:00:00 in local time).)

License Description Procedure before registration

Temporary license It is valid for 14 days from the time when starting Real-time Data Analyzer after

installed.

Page 28 Registration

procedure of a temporary

license

Subscription license It is valid in the specified period after issuing the license.

The expiration date can be checked in License Manager.

Page 28 Registration

procedure of a subscription

license

4 PROCEDURE BEFORE OPERATION4.1 License Manager 25

4

Startup method

Operating procedure

1. Select [Mitsubishi Electric Edgecross Software] [License Manager] from Windows Start.

The display language of License Manager is the same as the language set for the operating system.

Screen configuration

Window

(1) (2)


Displayed items

*1 It cannot be selected when "No file" is displayed in "State name".

Item Description

[License list] tab (1) Product name list A list of products subject to the license management and its status are displayed.

■Product name

Names of the products subject to the license management are displayed.

■State/State name

The license statuses of each product are displayed.

• "Success": Authentication succeeded

• "Failure": Authentication failed

• "Temporary": Temporary license in use

• "No file": No license file

(2) Detailed information The detailed information of a selected product is displayed.

■Entitlement ID

The entitlement ID of the subscription license in use is displayed.

'-' is displayed in any of the following cases.

• The temporary license is used.

• The entitlement ID cannot be displayed (In a case such as when an incorrect license file is

registered or a license file is not included etc.)

■License expiration

The expiration date of either the temporary license or subscription license is displayed. In a case such

as when an incorrect license file is registered or a license file is not included, the license expiration

date cannot be displayed and '-' is displayed instead.

• Temporary license

It is displayed as 'Days Remaining: {XX}' (XX: 1 to 14)

If the expiration date is less than one day, it will be displayed as 'Less than 1day'.

• Subscription license

The expiration date (local time) is displayed in the format 'YYYY/MM/DD hh:mm'.

■Details of status

When "Failure" or "No file" is displayed in "State name", the detailed information is displayed.

• "Failure" is displayed in "State name":

The error code and its detailed information are displayed. (Page 27 Error description)

• "No file" is displayed in "State name":

The message "A license file is not registered." is displayed.

When "Success" or "Temporary" is displayed in "State name", '-' is displayed.

[Registration] button Click this to register the license file of a selected product at the first time or update a registered license

file to new one.

Select a license file in the "Open" screen which appears by clicking this button.

[Delete] button*1 Click this to delete the registered license file of a selected product or migrate Real-time Data Analyzer

to another industrial PC.

[Information for activating]

tab

Hardware code The code created based on the hardware-specific information necessary for application of a license

file is displayed.

4 PROCEDURE BEFORE OPERATION4.1 License Manager 27

4

Error descriptionThe following table shows the error descriptions displayed in "Detail" and the corrective actions when "Failure" is displayed in

"State name".

Error code Error description Cause Corrective action

210018 There are problems with the license

file that is registered.

• The license file is corrupted.

• A license file issued for another

industrial PC is specified.

• The license file was replaced with

one other than an Edgecross Basic

Software license file.

Register the correct license file issued for the

industrial PC in use.

After registering a license file for the

first time to the industrial PC in use,

the date and time was changed and

the time on Windows was shifted

before the license issue date and

time.

Change the time on Windows to the time after the

license issue date and time.

• An Ethernet port or wireless LAN of

the industrial PC is disabled.

• An external Ethernet adapter in

use is removed.

• Enable the Ethernet port or wireless LAN of the

industrial PC.

• Connect the external Ethernet adapter.

210026 The Windows time has changed. After registering a license file for the

first time to the industrial PC in use,

the date and time was changed and

the time on Windows was shifted

forward or backward for 24 hours or

more.

Please consult your local Mitsubishi

representative.

210037 The license has expired. The temporary license expired. Please contact your local Mitsubishi

representative and acquire a new license, then

register the license.214109 The subscription license expired.

Other error codes A system error has occurred. Errors that seems not to occur in

actual operation occurred due to an

internal operation of the software.

Please contact your local Mitsubishi

representative and acquire a new license, then

register the license.


Registration procedure of a temporary licenseThis section shows the procedure for registering a temporary license.

Operating procedure

1. Click the [Registration] button in the [License list] tab in the "License Manager" screen.

2. Select 'RDA_Temp.elic' stored in the installation destination folder of Real-time Data Analyzer, and click the [Open]

button.

Registration procedure of a subscription licenseThis section shows the procedure for registering a subscription license.

Operating procedure

1. Check the entitlement ID shown in the Entitlement ID Notification.

When renewing a license, check also the entitlement ID in use which is displayed in "Entitlement ID" in the [License list] tab in

the "License Manager" screen.

2. Check the hardware code displayed in "Hardware code" in the "Information for activating" tab in the "License Manager"

screen.

3. Access to the web form for issuing licenses, then enter the following information and apply for the license.

• Entitlement ID

• Entitlement ID in use (only for a license renewal)

• Hardware code

• User information (company name, customer name, phone number, address, e-mail address, and others)

For the web form for issuing licenses, refer to the Real-time Data Analyzer License Application Instructions.

4. Save the license file delivered to the entered e-mail address into a folder on the industrial PC in which Real-time Data

Analyzer is installed.

5. Click the [Registration] button in the [License list] tab in the "License Manager" screen.

6. Select the license file saved in step 4, and click the [Open] button.

Precautions

• When checking the hardware code, enable only the Ethernet port which is used all the time while Real-time Data Analyzer

is operating. License authentication may fail if the Ethernet port is disabled after obtaining and registering a license file*1.

*1 It is the same as when removing an external Ethernet adapter or disabling the wireless LAN.

• Be sure to enter the correct e-mail address carefully to receive a license file without fail.

• Do not change the license file extension (.elic).

• Use this product with a temporary license while waiting for a license issuance after the application. If a temporary license

expires before obtaining a license, please contact your local Mitsubishi representative to reissue a temporary license.

4 PROCEDURE BEFORE OPERATION4.2 Startup Method 29

4

4.2 Startup MethodThis section shows the startup method of Real-time Data Analyzer.

Similar Waveform Recognition Tool

Operating procedure

1. Select [Mitsubishi Electric Edgecross Software] [Real-time Data Analyzer] from Windows Start.

2. Double-click 'Similar Wave Recognition Tool'.

Real-time Statistic Diagnosis Tool

Operating procedure


2. Double-click 'Real-time Statistic Diagnosis Tool'.

4.3 Display Language SwitchingReal-time Data Analyzer supports multiple language, so the display language such as one on the menu can be switched by

using the utility for switching language.

Operating procedure

1. Close Real-time Statistic Diagnosis Tool (Reader, Analyzer, and Monitor) if it is being used.


3. Double-click 'RDA_Language-Switching'.

The utility for switching language is started.

The display language of the utility for switching language follows the language setting of the operating system.

4. Select a display language and click the [Apply] button.

Precautions

• Set both the display language of the operating system and the system locale according to the language setting of Real-time

Statistic Diagnosis Tool to be started. Otherwise, the correct language may not be applied or characters may be garbled in

Real-time Statistic Diagnosis Tool.

• When changing the display language, the following settings in Real-time Statistic Diagnosis Tool are initialized according to

the changed language so reset the settings as necessary. However, for the following settings in Analyzer and Monitor, the

contents previously set are applied because the settings are remained for each switching display language.

Reader: Folder definition (Page 129 Folder Definition)

Analyzer: Management (Page 384 Management), Data management setting (Page 390 Data Management

Setting), Waveform display setting (Page 391 Waveform Display Setting), Logic setting (Page 393 Logic Setting)

Monitor: Monitor setting (Page 419 Monitor Setting)

• When switching the display language, set the following folders, which are set in Reader of Real-time Statistic Diagnosis

Tool, different from ones previously used before language switching. Otherwise, characters may be garbled because the

data in the language corresponding to the display language is saved in the following folders.

Reader equipment data folder

RSD dataset folder

304 PROCEDURE BEFORE OPERATION4.3 Display Language Switching

MEMO

31

PA

RT

2

PART 2 Similar Waveform Recognition Tool

This part explains Similar Waveform Recognition Tool.

Page 32 OVERVIEW

Page 36 WAVEFORM LEARNING

Page 38 DATA DIAGNOSIS

Page 44 PROCEDURE TO USE Similar Waveform Recognition Tool

Page 62 CREATING REFERENCE WAVEFORM LEARNING DATA

Page 65 PUBLISHING DATA SETTING

Page 68 OPERATION SETTING

Page 75 RESPONSE DATA SETTING

Page 77 EXECUTION/MONITORING OF SIMILAR WAVEFORM RECOGNITION

Page 81 OUTPUTTING DIAGNOSIS RESULT FILE

Page 90 GX LogViewer COOPERATION

Page 92 CONSIDERATIONS FOR SIMILAR WAVEFORM RECOGNITION

Page 98 TROUBLESHOOTING BY SYMPTOM

325 OVERVIEW

5 OVERVIEW

Similar Waveform Recognition Tool is an edge application that consists of the following functions.

• A function for learning the given cyclic waveform to be a reference (reference waveform) automatically. Even a waveform of

which the cycle is not constant or a waveform in which multiple patterns are switched, their waveforms are learned as the

characteristic of the normal cyclic waveforms.

• A function for diagnosing the similarity between a waveform (inspection waveform) input as a diagnosis target and a

learned waveform at high-speed and calculating an index for representing similarity (similarity score)

• A function for notifying both Real-time Flow Manager and users of the diagnosis result calculated based on the similarity

score

The following processes are necessary to analyze waveform data in Similar Waveform Recognition Tool.

1. Data analysis phase

A diagnosis criterion (reference waveform learning data) for an inspection waveform used for data diagnosis is created based

on data collected by Real-time Flow Manager. (Page 36 WAVEFORM LEARNING)

In this phase, a data logging flow of Real-time Flow Manager is performed.

For a data logging flow, refer to the following manual.


2. Data diagnosis phase

An inspection waveform is monitored and a waveform different from the reference waveform, that is the similarity score is

lower than a threshold value, is notified to Real-time Flow Manager if found. (Page 38 DATA DIAGNOSIS)

In this phase, a data diagnosis flow of Real-time Flow Manager is performed.

For a data diagnosis flow, refer to the following manual.


(1) Reference waveform

(2) Inspection waveform

(1)

(2) (2)

5 OVERVIEW5.1 Home screen 33

5

5.1 Home screenThis screen is displayed when starting Similar Waveform Recognition Tool.

All setting items are displayed.

Window

Displayed items

■ Menu configuration

*1 An error occurs if a setting file is imported in Real-time Data Analyzer the version of which is older than the one used for exporting the file.

*2 Up to 200 characters can be set in the file path of a file name.

■ Screen configuration

Menu item Description Reference

File Import Settings To load a setting file exported by using the

[Export Settings] menu to Similar Waveform

Recognition Tool.*1

Page 34 Setting file import

Export Settings To save the setting contents edited in Similar

Waveform Recognition Tool as a setting file.*2Page 34 Setting file export

Initialize All Settings To initialize the contents set after starting

Similar Waveform Recognition Tool to those

at the start-up.

Exit To exit Similar Waveform Recognition Tool.

Option Edgecross Link Setting To set parameters for cooperating with Real-

time Flow Manager in MQTT communication.

Edgecross Link Setting

Help Manual To start e-Manual Viewer and display the

manual.

Version Information To display the version information of Real-

time Data Analyzer.

Item Description Reference

[Creation of Reference Waveform

Learning Data] button

Click this to create reference waveform learning data. Page 62 CREATING REFERENCE WAVEFORM

LEARNING DATA

[Publishing Data Setting] button Click this to set publishing data. Page 65 PUBLISHING DATA SETTING

[Operation Settings] button Click this to set the operation setting for similarity diagnosis. Page 68 OPERATION SETTING

[Response Data Setting] button Click this to set response data. Page 75 RESPONSE DATA SETTING

[Execution/Monitoring] button Click this to start or stop a similarity diagnosis and display both

the execution status of each operation setting and the

waveform monitor.

Page 77 EXECUTION/MONITORING OF SIMILAR

WAVEFORM RECOGNITION

345 OVERVIEW5.1 Home screen

Setting file importThe following shows the procedure and considerations for importing a setting file.

ProcedureThe procedure for importing a setting file is as follows.

1. Install Real-time Data Analyzer in a import destination industrial PC.

This step is not necessary if importing a setting file in the same industrial PC as the one used at the time of export.

2. Select [File] [Import Settings] to import the setting file.

3. If any of the steps in the following section is not performed before exporting the setting file, perform them.

This step is not necessary if all the steps are performed before the export.

Page 44 PROCEDURE TO USE Similar Waveform Recognition Tool

If the settings of Real-time Statistic Diagnosis Tool have been configured, its current settings will be

overwritten by importing a setting file. It is recommended to export the current settings for a backup before

importing the file.

ConsiderationsConsiderations for importing a setting file are as follows.

• If any of the settings (example: IP address) in the "Edge Application Diagnosis Setting" screen of Real-time Flow Designer

is changed after importing a setting file, perform the operations in steps 2 and later described in the following:

Page 50 Data Diagnosis Setting

• To operate Similar Waveform Recognition Tool with the settings of when exporting a setting file even if the file and folder

configuration is different from the one at the time of the export, copy the corresponding files and folders after importing a

setting file. Files and folders that are required to be copied are shown in the following table.

*1 It is not required to be copied when not performing a waveform learning.*2 It is not required to be copied when not performing a similarity diagnosis.*3 It is not required to be copied when satisfying any of the following conditions:

A response data definition file has already been read in the "Edge Application Diagnosis Setting" screen of Real-time Flow Designer.A response data definition file will be output after importing a setting file.

Setting file exportThe following shows the considerations for exporting a setting file.

ConsiderationsConsiderations for exporting a setting file are as follows.

• Export targets are the setting contents in each screen only; files and folders themselves are not exported. If there is a

possibility to delete or move any of the files or folders, it is recommended to back them up when exporting a setting file.

Screen Item File/folder required to be copied

Reference Waveform Learning Data File

Creation setting

(Page 63 Reference Waveform

Learning Data File Creation Setting

Screen)

Folder to save the Reference Waveform File • Folder set for this item*1

• File stored in the set folder*1

Output folder Folder set for this item*1

Response Data Setting

(Page 75 RESPONSE DATA

SETTING)

[Response data definition file output] button Response data definition file output with the settings at the

time of export*2,*3

Operation Settings

(Page 69 Operation Setting Screen)

Reference waveform learning data file • File set for this item*2

• Folder in which a setting file is stored*2

5 OVERVIEW5.1 Home screen 35

5

Edgecross Link SettingSet parameters for cooperating with Real-time Flow Manager in MQTT communication.

Window

Select [Option] [Edgecross Link Setting].

Displayed items

Item Description Consideration

Connection Address The IP address of an MQTT broker of a connection destination is displayed.

The value is fixed to '127.0.0.1' (localhost).

Port number Set the port number (1024 to 65535). Set the same port number as one set

for an MQTT broker used.

User name Set a user name (up to 32 characters). Set the same user name as one set for

an MQTT broker used.

Password Set a password (up to 32 characters). Set the same password as one set for

an MQTT broker used.

[Apply] button Click this to apply the set contents.

[OK] button Click this to apply the set contents and close the screen.

366 WAVEFORM LEARNING6.1 Operation Overview in Waveform Learning

6 WAVEFORM LEARNING

The waveform learning function creates a diagnosis criterion (reference waveform learning data) of an inspection waveform

used for data diagnosis based on data collected by Real-time Flow Manager.

Saving waveform data collected by Real-time Flow Manager

Learning based on the reference waveform file

Outputting the learning data to a file

6.1 Operation Overview in Waveform LearningThis section shows the operation overview in waveform learning.

The partial waveform group is created by extracting waveforms for the number of records of the unit waveform width from a

reference waveform group in a learning period while shifting the waveform by one record.

The partial waveform group is used as an index to see how similar a waveform is to the inspection waveform when performing

a data diagnosis flow.

Detailed operations in waveform learning are as follows:

Importing a waveform for the period specified as the learning period

Analyzing and learning the shape of waveforms (partial waveform) for the number of records specified in the unit waveform

width

Repeating for each record (collection cycle)

Completing both learning all partial waveforms in a learning period and saving them in a reference waveform learning data

file, and ending the analysis.


(2) Similar waveform recognition (waveform learning)

(a) Reference waveform file

(b) Reference waveform learning data file

(1) Learning period

(2) Collection interval

(3) Unit waveform width

(4) Partial waveform group

(a)�

(b)

(2)(1)

�

�

�

�

��

�

(2)

(3)

(1)

(4)

6 WAVEFORM LEARNING6.1 Operation Overview in Waveform Learning 37

6

MEMO

387 DATA DIAGNOSIS

7 DATA DIAGNOSIS

The data diagnosis function monitors an inspection waveform and notifies Real-time Flow Manager of the number of

waveforms different from the reference waveform, that is the similarity score is lower than a threshold value, and the gap

between a similarity score and similarity score threshold acquired in each individual similarity diagnosis.

In addition, a similarity diagnosis can be paused by monitoring publishing data on the specified conditions. (Page 43

Pause of Similarity Diagnosis)

An inspection waveform are displayed in real time by connecting to GX LogViewer. (Page 90 Similar waveform

recognition monitor)

A diagnosed inspection waveform and the diagnosis result data is saved in a file (outputting the diagnosis result file) and the

diagnosis status can be checked after the diagnosis.

For a diagnosis result file, refer to the following:


For the display of a diagnosis result file in GX LogViewer, refer to the following:

Page 91 Diagnosis result file display

• GX LogViewer is installed at the same time when installing Real-time Data Analyzer.

• When the load on an industrial PC is high, the limit of the processing performance may be exceeded and

the similarity diagnosis may stop. Before starting an actual operation, monitor the load of the PC and check

that a similarity diagnosis does not stop by following the procedure described in the following section.

Page 54 Starting a Temporary Operation

7 DATA DIAGNOSIS 39

7

Acquiring the information, such as the name and the data type, of an inspection waveform to receive in MQTT

communication by using the publishing data definition file*1

*1 It can be used only the publishing data definition file which was output in the edge application diagnosis (MQTT) setting of Real-time Flow Designer.

Comparing the reference waveform with an inspection waveform and evaluating the similarity (Calculating the similarity

score)

Acquiring the information, such as the name and the data type, of a diagnosis result to receive in MQTT communication by

using the response data definition file

Inputting an inspection waveform

Outputting the diagnosis result data

Saving both the diagnosed inspection waveform and the diagnosis result data in a file

Reporting the diagnosis result in real time

Checking the diagnosis status by using the diagnosis result file


(2) MQTT broker

(3) Similar waveform recognition (similarity diagnosis)

(4) GX LogViewer

(5) GX LogViewer or others

(a) Publishing data definition file

(b) Reference waveform learning data file

(c) Response data definition file

(d) Diagnosis result file

(2)

��

(b)

(3)(1)

(a)

(c)

��

��

(4)

��

(d)(5)

��

��

��

� ��

��

Offline

Online

407 DATA DIAGNOSIS7.1 Operation Overview in Similarity Diagnosis

7.1 Operation Overview in Similarity DiagnosisThis section shows the operation overview in similarity diagnosis.

When performing a data diagnosis flow, the partial waveform group created in waveform learning and a waveform acquired by

extracting an inspection waveform into a unit waveform width are compared and the similarity score of each is calculated.

Among the multiple similarity scores, the highest one will be the similarity score as the diagnosis result.

When the calculated similarity score is higher than the similarity score threshold corresponding to the detection sensitivity, the

waveform is determined to be similar. If the score is lower, the waveform is determined to be different.

Detailed operations in similarity diagnosis are as follows:

Collecting data every collection interval

Extracting data for the unit waveform width

Comparing the partial waveform group saved in waveform learning with the waveform extracted in , and calculating the

similarity score of each partial waveform

Determining the waveform whether it is similar or different. It is determined to be similar when the highest similarity score

calculated in is higher than the similarity score threshold corresponding to the detection sensitivity, and it is determined to

be different if the score is lower. (The highest similarity score will be the similarity score as the diagnosis result.)

Shifting the unit waveform width backward by the number of records specified in the data diagnosis cycle

Repeating to

Ex.

When the similarity score threshold corresponding to the detection sensitivity is 65:

: It is determined as a normal waveform because the highest similarity score '92' is higher than the similarity score threshold '65' corresponding to the

detection sensitivity.

: It is determined as an abnormal waveform because the highest similarity score '58' is lower than the similarity score threshold '65' corresponding to the

detection sensitivity.

(1) Collection interval

(2) Unit waveform width

(3) Highest similarity score

(4) Most similar partial waveform

(5) Partial waveform group

(1)

(2)(3)

(4)

(5)

92

58

�

�

�

�

��

�

7 DATA DIAGNOSIS7.2 Similarity Score 41

7

7.2 Similarity ScoreThe similarity score is an index value that indicates how similar an inspection waveform and a reference waveform are.

The value of the similarity score is in the range from 0 to 100. When a reference waveform and an inspection waveform is

completely matched, the similarity score will be '100'. When an inspection waveform is surely determined as abnormal

comparing to a reference waveform, the similarity score will be '0'.

However, due to the effects of the optimization processing for high-speed calculation or slight fluctuation on the time axis of

collection data, the similarity score may not be 100 even if an inspection waveform and a reference waveform are completely

matched. (In this case, the similarity score marks a high score such as '99' or '98'.)

• Comparison of waveforms to be completely matched (Similarity score = 100)

• Comparison of waveforms to be possibly determined as abnormal (Similarity score = 50)

• Comparison of waveforms to be surely determined as abnormal (Similarity score = 0)

Precautions

• Even if a reference waveform and an inspection waveform are smooth (fluctuation is small), the similarity score fluctuation

may increase because of the effects of the optimization processing for high-speed calculation or the jitter (slight fluctuation

on the time axis) of collection data.

• Even if the shapes of a reference waveform and an inspection waveform are completely matched, the similarity score may

be extremely low because of the slight fluctuation of the data collection timing.

• Set the detection sensitivity considering the above contents.







(1)

(2)

(1)

(2)

(1)

(2)

427 DATA DIAGNOSIS7.3 Similarity Score Threshold List

7.3 Similarity Score Threshold ListThe similarity score threshold list is the index data to determine the appropriate detection sensitivity by using the value of a

similarity score corresponding to the inspection sensitivity (1 to 6) calculated in waveform learning.

The values of the similarity score threshold list differ depending on the property of the reference waveform. The more similar

the shapes of the periodic waveforms included in the reference waveform, the values of the similarity score threshold list tend

to increase.

When the shape of each periodic waveform included in the reference waveform has no fluctuations or its fluctuations are too

small (triangle wave, for example), the similarity score may become 100; this increases false detection even with little

deviation. Decrease the sensitivity for detection if a waveform that should not be detected is incorrectly detected.

• Waveform that the similarity score threshold is high

• Waveform that the similarity score threshold is low

• Waveform without fluctuations or one whose fluctuations are too small

Reference waveform Similarity score threshold list

Detection sensitivity Similarity score

6 97

5 92

4 86

3 80

2 73

1 66



6 82

5 76

4 71

3 66

2 60

1 55



6 100

5 100

4 100

3 100

2 100

1 50

7 DATA DIAGNOSIS7.4 Pause of Similarity Diagnosis 43

7

7.4 Pause of Similarity DiagnosisA similarity diagnosis can be paused by monitoring publishing data on the specified conditions.

By pausing the diagnosis when an error such as an emergency stop of equipment occurs, the diagnosis result is not

determined to be different and an expected result can be obtained.

In addition, up to eight units of publishing data can be monitored on the specified conditions per publishing data definition file.

The diagnosis can be paused when any or all of the conditions are satisfied.

Available comparison operators and constant values to compare for each type of publishing data are as follows.

*1 When the data type is REAL or LREAL, the result may not be determined correctly due to a floating point error.*2 The following strings are excluded:

U+0000 to U+001F, and U+007F*3 The following strings are excluded:

U+0000 to U+001F, U+007F to U+00A0, U+00AD, U+2028, and U+2029

Data type Operator Constant value to compare

BOOL =, TRUE, FALSE

INT =, , <, , >, -32768 to 32767 (integer)

UINT 0 to 65535 (integer)

DINT -2147483648 to 2147483647 (integer)

UDINT 0 to 4294967295 (integer)

REAL*1 • 0

• 1.17550E - 38 (absolute value) to 3.40282E + 38 (absolute value)

LREAL*1 • 0

• 2.22507385850721E - 308 (absolute value) to 1.79769313486231E + 308

(absolute value)

STRING =, String (ASCII)*2

WSTRING =, String (Unicode)*3

448 PROCEDURE TO USE Similar Waveform Recognition Tool

8 PROCEDURE TO USE Similar Waveform Recognition Tool

This chapter shows the diagnosis procedure by cooperating Similar Waveform Recognition Tool and Real-time Flow Manager.

Since Similar Waveform Recognition Tool can cooperate with Real-time Flow Manager that is installed in another industrial

PC, data does not need to be collected or diagnosed in a same industrial PC; therefore, load on the industrial PC can be

reduced.

Operating procedure

1. Create a diagnosis criterion (reference waveform learning data) for an inspection waveform used for data diagnosis

based on data collected in a data logging flow of Real-time Flow Manager.

Page 45 Data Analysis Setting

2. Create both the operation setting for similarity diagnosis by using the created reference waveform learning data file and

the data diagnosis flow setting by using Real-time Flow Designer.


3. Perform a data diagnosis flow and adjust both the detection sensitivity and data diagnosis cycle.

Page 54 Starting a Temporary Operation

4. Start an actual operation.

Page 57 Starting an Actual Operation

Precautions

• When logging in the industrial PC by using the remote log-in function of Windows or unlocking a screen on the PC while

Similar Waveform Recognition Tool is running, the error message 'Loss of the publishing data received from the Edgecross

side is found.' appears and the tool may stop operating*1. Even when the error message does not appear, a diagnosis may

not performed in real time or distribution of a diagnosis result to Real-time Flow Manager may be delayed.

Therefore, it is recommended to use the tool after logging in by using the remote log-in function of Windows or setting not to

unlock screens.

*1 When the operation is stopped, switch the operating status of Real-time Flow Manager to STOP , then to RUN again.

• When cooperating with Real-time Flow Manager that is installed in another industrial PC, a communication error may occur

due to network disconnection or others. If an expected diagnosis result cannot be obtained, check if a communication error

has occurred in the diagnostic screen of Real-time Flow Designer.

8 PROCEDURE TO USE Similar Waveform Recognition Tool8.1 Data Analysis Setting 45

8

8.1 Data Analysis SettingThis section shows the procedure for creating a diagnosis criterion (reference waveform learning data) for an inspection

waveform used for data diagnosis based on data collected in the data logging flow of Real-time Flow Manager.

Operating procedure

1. Create a data logging flow setting by using Real-time Flow Designer.

Page 46 Creating the data logging flow setting

2. Perform a data logging flow with the created data logging flow setting and acquire a reference waveform.

Page 49 Acquiring a reference waveform

3. Create reference waveform learning data by using Similar Waveform Recognition Tool.

Page 49 Creating reference waveform learning data

468 PROCEDURE TO USE Similar Waveform Recognition Tool8.1 Data Analysis Setting

Creating the data logging flow settingA data logging flow setting can be created by using Real-time Flow Designer.

For details on each setting, refer to the following:


• A diagnosis in a similar waveform recognition is suitable for data in which dates and times are arranged at

regular intervals in chronological order, so set a collection interval to be constant.

• Set the setting so as not to occur an error such as the collection cycle exceeded event. (Page 424

Similar Waveform Recognition Tool)

• Include the data storing setting (file saving) in a data logging flow setting in order to create a CSV file

(reference waveform file) including a reference waveform used for waveform learning.

Operating procedure

1. Start Real-time Flow Designer.

2. Click "Data Logging Flow Setting" in the edit item tree.

3. Click the [Edit] button.

The "Data Logging Flow Setting" screen appears.

4. Set the setting for data collection.

The following table shows the considerations for the setting.

*1 It can be used to pause a similarity diagnosis. (Page 43 Pause of Similarity Diagnosis)

Item Consideration

Data collector setting • It is recommended to use a data collector that reduces variations of

collected data.

Collection data setting The following types of data are not available for waveform learning and similar

waveform recognition.

• BOOL type*1

• LINT type

• ULINT type

• STRING type*1

• WSTRING type*1

Collection interval Set a collection interval so that change of a waveform to be collected can be

detected.


8

5. Set the data modification setting as necessary.

With data modification, some dates and times may not be continuous in chronological order at regular interval. Make sure that

the number of units of continuous data in which dates and times are arranged at regular intervals in chronological order is the

number of records four times or more of the unit waveform width at least.

Ex.

When modifying data to extract values which are less than a certain condition and if a section in which the dates and times are

continuous are less than four times of a unit waveform width, a similar waveform can not be detected and it is determined as

abnormal.

6. Set the data storing setting.


(1) Collected data

(2) Condition value

(3) Data extraction (values less than the condition value are extracted.)

(4) Arrangement in which data is continuous

(5) Modified data

(6) Section in which the dates and times are continuous

Item Consideration

Executable form Select "File".

Output data setting The following types of data are not available for waveform learning and similar waveform recognition.

• BOOL type*1

• LINT type

• ULINT type

• STRING type*1

• WSTRING type*1

In the following cases, waveform learning and similar waveform recognition cannot be performed with Similar

Waveform Recognition Tool in Real-time Data Analyzer whose version is 1.00A.

• The data type is REAL and the number of digits after the decimal point is set to 6 or more.

• The data type is LREAL and the number of digits after the decimal point is set to 15 or more.

Date and time format Output the date and time which allows for representing the minimum unit of a collection interval.

■Example

• When the collection interval is 100 ms: Representation down to hundreds place of millisecond is necessary

("YYYY/MM/DD hh:mm:ss.s").

• When the collection interval is 5 ms: Representation down to ones place of millisecond is necessary ("YYYY/

MM/DD hh:mm:ss.sss").

When any of the following reserved words is used, waveform learning and similar waveform recognition cannot be

performed with Similar Waveform Recognition Tool in Real-time Data Analyzer whose version is 1.00A.

• '.sssss'

• '.ssssss'

• '.sssssss'

• '.ssssssss'

• '.sssssssss'

• '.ns'

(1)(2)

(3)

(5)

(6)

(4)

488 PROCEDURE TO USE Similar Waveform Recognition Tool8.1 Data Analysis Setting



8

Acquiring a reference waveformAccording to the created data logging flow setting, acquire a reference waveform by executing the data logging flow.

For performing a data logging flow, refer to the following:


To improve the diagnosis precision, refer to the following section and take a necessary action.

Page 92 Waveforms Used for Similar Waveform Recognition

Page 94 Input Waveform (Reference Waveform) and Setting Items in Waveform Learning

• For similar waveform recognition, depending on the properties of waveforms, some are suitable for

diagnosis but others are not. Therefore, it is recommended to select waveforms supposed to be a reference

waveform by checking them in advance. (Page 92 Waveforms Used for Similar Waveform Recognition)

• If fluctuation or deviation is large in a waveform which is in the normal state, more accurate reference

waveform learning data can be created by increasing the number of records to be acquired. The number of

records to be a reference depends on the property of a waveform so adjust it in a temporary operation.

(Page 54 Starting a Temporary Operation)

Precautions

• Acquire a reference waveform in which no data is missing. Otherwise, diagnosis may not performed properly.

• For a reference waveform, it is necessary to collect data to include waveforms for at least four cycles*1 of the periodic

waveform to be input. When only waveforms for four cycles are used for waveform learning, expected diagnosis accuracy

is rarely acquired. Therefore, it is recommended to acquire as much data as possible within the creation time. (Page

425 Industrial PC (operation-guaranteed environment))

*1 If a long-cycle ripple is included in a waveform to be collected, take into consideration for the cycle of such a waveform and acquire data.

Creating reference waveform learning dataCreate reference waveform learning data by using Similar Waveform Recognition Tool. (Page 62 CREATING

REFERENCE WAVEFORM LEARNING DATA)

To improve the diagnosis precision, refer to the following section and take a necessary action.


Page 94 Input Waveform (Reference Waveform) and Setting Items in Waveform Learning

(1) Acquired data

(2) Separating for each waveform component

(3) Principal component

(4) Ripple component

(5) Waveform for 4 cycles

(1)

(2)

(3) (4)

(5) (5)

508 PROCEDURE TO USE Similar Waveform Recognition Tool8.2 Data Diagnosis Setting

8.2 Data Diagnosis SettingThis section shows the procedure for creating both the operation setting for similarity diagnosis by using the created reference

waveform learning data file and the data diagnosis flow setting by using Real-time Flow Designer.

Operating procedure

1. Set up an MQTT broker in an industrial PC in which Similar Waveform Recognition Tool is installed.

For the setup of an MQTT broker, refer to the following:


2. Create a publishing data definition file by using Real-time Flow Designer.

Page 51 Creating a publishing data definition file

3. Set the settings for cooperating with Real-time Flow Manager in MQTT communication by using Similar Waveform

Recognition Tool.

Page 35 Edgecross Link Setting

4. Set the publishing data by using Similar Waveform Recognition Tool.


5. Set the operation setting for similarity diagnosis by using Similar Waveform Recognition Tool.


6. Set the response data and output a response data definition file by using Similar Waveform Recognition Tool.


7. Read the response data definition file by using Real-time Flow Designer.

Page 53 Reading a response data definition file

8. Create a data diagnosis flow by using Real-time Flow Designer.

Page 53 Creating the data diagnosis flow setting

To create multiple data diagnosis flows, repeat from step 2 to 8 for each data diagnosis flow. However, step 3

can be skipped if already set once.

Precautions

• Do not set one publishing data definition file to multiple publishing data settings. An error occurs when setting the publishing

data.

• In the loading of a response data definition file, do not specify the same file for multiple data diagnosis flow setting.

Otherwise, input and output data will be inconsistency between an edge application and Real-time Flow Manager and a

diagnosis will not be performed properly.

• If data missing occurs in the collected data, it may be erroneously diagnosed that there is a difference in a waveform.

Therefore, set the settings so that no data missing occurs in collected data.

• If there is data including a surrogate pair character in a publishing data definition file, the file cannot be read with Similar

Waveform Recognition Tool in Real-time Data Analyzer whose version is 1.02C or earlier.

8 PROCEDURE TO USE Similar Waveform Recognition Tool8.2 Data Diagnosis Setting 51

8

Creating a publishing data definition fileA publishing data definition file can be created by using Real-time Flow Designer.



Operating procedure


2. Click "Data Diagnosis Flow Setting" in the edit item tree.

3. Click the [Move from Data Logging Flow] button.

4. Select a flow set in "Data Logging Flow Setting" and click the [] button.

5. Click the [OK] button.

The flow set in "Data Logging Flow Setting" moves to "Data Diagnosis Flow Setting".


The "Data Diagnosis Flow Setting" screen appears.

7. Select "Edge Application Diagnosis (MQTT)" in the "Function Type" column of a data diagnosis, then click the cell of the

"Detailed Setting" column.

528 PROCEDURE TO USE Similar Waveform Recognition Tool8.2 Data Diagnosis Setting

8. Set each item in the [Destination] tab in an edge application diagnosis (MQTT) setting.


9. Set each item in the [Output Data] tab in an edge application diagnosis (MQTT) setting.



10.Set each item in the [Definition output] tab in an edge application diagnosis (MQTT) setting, then click the [Publishing

Data Definition File output] button.


Precautions

• After outputting a publishing data definition file, do not operate the screen until the following procedure is completed.


• The name of a publishing data definition file should indicate that it is used in Similar Waveform Recognition Tool and which

data diagnostic flow corresponds to the file.

• When changing the data diagnosis flow setting that was set between the start and the end of output of a publishing data

definition file, initialize the operation setting of similarity diagnosis or delete the setting related to the publishing data setting

corresponding to the data diagnosis flow setting in the "Operation Settings" screen. Then, re-create the publishing data

definition file.


Item Consideration

Host name/IP address Set the same host name or IP address as the one for an industrial PC in which

Similar Waveform Recognition Tool is installed.

Port number Set the same port number as one set for an MQTT broker used. Otherwise, an

MQTT communication to Similar Waveform Recognition Tool fails and a

diagnosis is not performed.

User authentication information (user name and password) Set the same user authentication information, user name and password, as

one set for an MQTT broker used.

Communication encryption setting Do not use this setting because Similar Waveform Recognition Tool does not

support the encryption of MQTT communication. If used, an MQTT

communication to Similar Waveform Recognition Tool fails and a diagnosis is

not performed.

Client ID setting Do not set it in Similar Waveform Recognition Tool.

Item Consideration

Output data setting The following types of data are not available for waveform learning and similar

waveform recognition.

• BOOL type*1

• LINT type

• ULINT type

• STRING type*1

• WSTRING type*1

Item Consideration

Confirm existence of the Edge application at the time of delivery Select this checkbox to detect that a similarity diagnosis stops in Real-time


Topic name specification Do not set it in Similar Waveform Recognition Tool.

Client ID setting Do not set it in Similar Waveform Recognition Tool.

8 PROCEDURE TO USE Similar Waveform Recognition Tool8.2 Data Diagnosis Setting 53

8

Reading a response data definition fileA response data definition file can be read by using Real-time Flow Designer.



Operating procedure

1. Click the [Get Response Data definition] button in the [Response Data] tab in an edge application diagnosis (MQTT)

setting.

2. Select the response data definition file which was out put in the following setting.


An error occurs in Real-time Flow Manager if a different file is specified.

Select the response data definition file which was output according to the publishing data definition file loaded in the

publishing data setting described in the following section (The name set in the setting is included in the response data

definition file). If a wrong file is selected, a diagnosis will not be performed properly.


3. Click the [OK] button without changing the output data setting name.

Creating the data diagnosis flow settingA data diagnosis flow setting can be created by using Real-time Flow Designer.



Operating procedure

1. Set the feedback setting.


548 PROCEDURE TO USE Similar Waveform Recognition Tool8.3 Starting a Temporary Operation

8.3 Starting a Temporary OperationThis section shows the procedure for performing a data diagnosis flow and adjusting both the detection sensitivity and data

diagnosis cycle.

Operating procedure

1. Start an MQTT broker.

For the startup method of an MQTT broker, refer to the following:


2. Perform a similarity diagnosis by using Similar Waveform Recognition Tool.


3. Switch the operating status of Real-time Flow Manager to RUN and perform a data diagnosis flow by using Real-time

Flow Designer.

For the method for switching the operating status of Real-time Flow Manager, refer to the following:


If the operating procedure from step 1 to 3 is not followed in the order, an error such as an MQTT transmission

data overflow error occurs in Real-time Flow Manager.

4. Monitor the "Similar Waveform Recognition execution status detailed monitoring" screen and the CPU usage rate, and

adjust both the detection sensitivity and data diagnosis cycle.

Page 79 Similar Waveform Recognition Execution Status Detailed Monitoring Screen

Page 55 Detection sensitivity adjustment

Data diagnosis cycle adjustment

• In a temporary operation, collect only the data determined to be similar.

• The optimum detection sensitivity differs depending on the stability of a reference waveform (magnitude of

fluctuation and/or amount of noise).

• If similarity is diagnosed when the load on the industrial PC is high, the diagnosis may stop or data missing

may occur in the collected data and it may be erroneously diagnosed that there is a difference in a

waveform. Adjust the data diagnosis cycle and the load to other applications so that the load on the PC is

not too high.

• When configuring the data diagnosis setting, the detection sensitivity can also be adjusted either by

adjusting its level with the slider while checking the similarity score threshold displayed in the "Operation

Setting" screen or by entering the similarity score threshold directly in the screen. (Page 69 Operation

Setting Screen)

After starting a temporary operation, adjust the detective sensitivity by checking the "Similar Waveform

Recognition execution status detailed monitoring" screen. (Page 55 Detection sensitivity adjustment)

5. To stop the diagnosis function of Real-time Flow Manager, follow step 6 and 7 shown below to stop the data diagnosis

flow.

6. Switch the operating status of Real-time Flow Manager to STOP by using Real-time Flow Designer.

7. Perform the similarity diagnosis by using Similar Waveform Recognition Tool.

If the operating procedure from step 6 to 7 is not followed in the order, an error such as an MQTT transmission

data overflow error occurs in Real-time Flow Manager.

8 PROCEDURE TO USE Similar Waveform Recognition Tool8.3 Starting a Temporary Operation 55

8

Detection sensitivity adjustmentPerform the following procedure to determine the detection sensitivity.

1. Perform a data diagnosis flow until the lowest similarity score is stable, and then check the lowest value.

2. Check the similarity score threshold list.

3. Compare the lowest similarity score with the values in the similarity score threshold list, and then determine the detection

sensitivity that corresponds to the similarity score threshold list not exceeding the lowest similarity score.

For details on the similarity score and similarity score threshold list, refer to the following:

Page 41 Similarity Score

Page 42 Similarity Score Threshold List

■ Example of detection sensitivity adjustment • When the amplitude of the similarity score is small (stable waveform data that is small fluctuation and little noise):

If the following result is acquired by executing a data diagnosis flow for a certain period, set the detection sensitivity to 5.

• When the amplitude of the similarity score is large (unstable waveform data that is large fluctuation and much noise):

If the following result is acquired by executing a data diagnosis flow for a certain period, set the detection sensitivity to 3.

(1) Lowest similarity score = 95

(2) Detection sensitivity

(3) Similarity score threshold list

(1) Lowest similarity score = 68



6

5

4(2)

(1)

(3)3

2

1

97

92

86

80

73

66

(1)

6

5

4(2) (3)

3

2

1

82

76

71

66

60

55

568 PROCEDURE TO USE Similar Waveform Recognition Tool8.3 Starting a Temporary Operation

Precautions

• The example of detection sensitivity adjustment is a guide only. In an actual operation, adjust the detection sensitivity in a

user's discretion.

• The optimum detection sensitivity differs depending on the stability of a reference waveform (magnitude of fluctuation and/

or amount of noise).

• When the lowest similarity score is lower than a value of the similarity score threshold list corresponding to detection

sensitivity 1, review the procedure because the following contents are considered.

Data diagnosis cycle adjustmentFluctuate the data diagnosis cycle in the range from the lower and upper limits shown below to determine the cycle in which

the load on CPU is not too high.

• Lower limit value: Quarter of the unit waveform width or a value calculated by '100 collection interval (msec)', whichever is

smaller

• Recommended value: Quarter of the unit waveform width

• Upper limit value: Half of the unit waveform width

Precautions

The detection precision will be decreased when the data diagnosis cycle which is greater than the half of a unit waveform

width is set.

Check point Review method

The reference waveform is waveform data unsuitable for similar waveform

recognition.

Refer to the following section and check whether the reference waveform is

waveform data suitable for similar waveform recognition.


The date and time interval of the reference waveform and that of the

inspection waveform are not matched.

Check whether the collection interval set in the data logging flow setting is the

same as that of the data diagnostic flow setting.



The number of records of the reference waveform is too few. Refer to the following section and increase the number of records of the

reference waveform to be acquired.


Refer to the following section and extend the learning period.


Data different from the reference waveform is set to an inspection waveform. Check whether data to be collected is the same between the data logging flow

setting and the data diagnostic flow setting.



8 PROCEDURE TO USE Similar Waveform Recognition Tool8.4 Starting an Actual Operation 57

8

8.4 Starting an Actual OperationAn actual operation is started.

Precautions

If an expected diagnosis result cannot be obtained or any other problems occur after starting the operation, refer to the

following:


Checking the diagnosis status after starting an actual operation

Monitoring the similarity diagnosis status and the diagnosis informationBy using Similar Waveform Recognition Tool, the similarity diagnosis status (such as the operating status and diagnosis

status) and the diagnosis information (such as a similarity score and the number of times of difference detections) can be

monitored to check whether a similarity diagnosis is performed properly.


Monitoring the graphs of an inspection waveform and the similarity scoreBy using GX LogViewer, an inspection waveform and the similarity score can be monitored.

Page 90 Similar waveform recognition monitor

Checking a waveform when a difference is detectedBy using GX LogViewer, each waveform graph when a difference is detected can be checked.


To do so, select "Only difference condition" for "Output Settings" in "Diagnosis Result file saving setting."

Page 69 Operation Setting Screen



Checking the result of a similarity diagnosisIn a diagnosis result file, the result of a similarity diagnosis (whether an inspection waveform is similar to or different from a

reference waveform) can be checked.

To do so, select "Fixed cycle output" for "Output Settings" in "Diagnosis Result file saving setting."




588 PROCEDURE TO USE Similar Waveform Recognition Tool8.4 Starting an Actual Operation

Changing any settings after starting an actual operation

Changing the detection sensitivity setting or diagnosis result file saving settingThe following shows the procedure for restarting a diagnosis when changing the detection sensitivity setting or diagnosis

result file saving setting in the "Operation Settings" screen after starting an actual operation.

Operating procedure

1. Click the [Stop Operation] button in the "Real-time Flow Manager Diagnostics" screen of Real-time Flow Manager to stop

its operation.

For the method for displaying the "Real-time Flow Manager Diagnostics" screen, refer to the following:


2. Click the [Stop Operation] button in "Operation status" in the "Execution/Monitoring of Similar Waveform Recognition"

screen of Similar Waveform Recognition Tool to stop its operation.


3. Click the "Operation Settings" column for a setting to be changed in the "Operation Settings" screen of Similar Waveform

Recognition Tool.


4. Change the detection sensitivity setting or the diagnosis result file saving setting.


5. Click the [Start Operation] button in "Operation status" in the "Execution/Monitoring of Similar Waveform Recognition"

screen of Similar Waveform Recognition Tool to restart its operation.


6. Click the [Start Operation] button in the "Real-time Flow Manager Diagnostics" screen of Real-time Flow Manager to

restart its operation.

Precautions

• The diagnosis result file saving setting is set for each diagnosis; therefore, it is required to change the setting per diagnosis.

• After restarting a diagnosis, check the "Similar Waveform Recognition execution status detailed monitoring" screen and

adjust the detection sensitivity so that it corresponds to the similarity score threshold not exceeding the lowest similarity


8

score.

• To stop outputting diagnosis result files to move the files or change the diagnosis result file saving setting, click [Suppress]

button in "Diagnosis Result File output status (Fixed cycle output)" of the "Execution/Monitoring of Similar Waveform

Recognition" screen in Similar Waveform Recognition Tool. (To restart the output, click the [Cancel suppression] button.)


Changing a response data settingThe following shows the procedure for changing a response data setting after starting an actual operation.

Operating procedure

1. Click the [Stop Operation] button in the "Real-time Flow Manager Diagnostics" screen of Real-time Flow Manager to stop

its operation.

For the method for displaying the "Real-time Flow Manager Diagnostics" screen, refer to the following:


2. Click the [Stop Operation] button in "Operation status" in the "Execution/Monitoring of Similar Waveform Recognition"

screen of Similar Waveform Recognition Tool to stop its operation.


3. Click the "Response Data Setting" column for a setting to be changed in the "Response Data Setting List" screen of

Similar Waveform Recognition Tool.


4. Select response data to be output in the "Response Data List" screen.

Page 76 Response Data List Screen

608 PROCEDURE TO USE Similar Waveform Recognition Tool8.4 Starting an Actual Operation

5. Select the checkbox in the "Output" Column for a setting to output its response data definition file in the "Response Data

Setting List" screen, and click the [Response data definition file output] button.


6. Click "Data Diagnosis Flow Setting" in the edit item tree of Real-time Flow Designer.

The data diagnosis flow setting list is displayed.

7. Select a data diagnosis flow to be changed.

The "Data Diagnosis Flow Setting" screen opens.

8. Change the setting in the "Data Storing" column for the "Data Diagnosis" process type to "Do not execute" if it is set to

"Execute." In addition, if "Already set" is displayed in the "Detailed Setting" column for the "Feedback" process type,

change the function type to "No Processing."

9. Click [Already Set] in the "Detailed Setting" column for the "Data Diagnosis" process type in the "Data Diagnosis Flow

Setting" screen.

The "Edge Application Diagnosis Setting" screen appears.


8

10.Click the [Get Response Result Definition] button in the [Response Data] tab of the "Edge Application Diagnosis Setting"

screen to select the response data definition file.


12.Set the data diagnosis setting and feedback setting, and click the [OK] button.

13.Click the [Start Operation] button in "Operation status" in the "Execution/Monitoring of Similar Waveform Recognition"

screen of Similar Waveform Recognition Tool to restart its operation.


14.Click the [Start Operation] button in the "Real-time Flow Manager Diagnostics" screen of Real-time Flow Manager to

restart its operation.

Precautions

• When changing a response data setting in Similar Waveform Recognition Tool, it is required to change the setting in Real-

time Flow Designer as well. Therefore, back up the setting in advance.

• When changing a response data setting in Similar Waveform Recognition Tool, it is required to read the definition

information of the response data of which the setting is changed in Real-time Flow Designer. Otherwise, an error occurs in

Real-time Flow Designer.

629 CREATING REFERENCE WAVEFORM LEARNING DATA

9 CREATING REFERENCE WAVEFORM LEARNING DATA

This chapter shows the settings for creating reference waveform learning data.

Up to 64 units of reference waveform learning data can be created at once.

Window

Click the [Creation of Reference Waveform Learning Data] button.

Displayed items

By clicking the [Cancel] button in the screen displayed while creating reference waveform learning data, the

processing is canceled. However, the reference waveform learning data created before the cancellation is not

deleted.


(1) Setting list The list of reference waveform learning data file creation

settings is displayed.

File Name The file name of reference waveform learning data set in

the "Reference Waveform Learning Data File Creation

setting" screen is displayed.

Generation target setting The content set for "Generation target setting" in the

"Reference Waveform Learning Data File Creation

setting" screen is displayed.

Creation setting Click this to open the "Reference Waveform Learning

Data File Creation setting" screen.

Page 63 Reference Waveform

Learning Data File Creation

Setting Screen

[Delete] button Click this to delete the setting on the selected row.

When an error occurs while deleting multiple rows, the

settings on the rows before the error occurred will be

deleted.

[Start creation.] button Click this to start creating reference waveform learning

data.

(1)

9 CREATING REFERENCE WAVEFORM LEARNING DATA9.1 Reference Waveform Learning Data File Creation Setting Screen 63

99.1 Reference Waveform Learning Data File Creation

Setting ScreenThis section shows the detailed setting for creating reference waveform learning data.

Window

Click the "Creation setting" column in the setting list.

Displayed items


Folder to save the Reference Waveform

File*1Set the folder path containing a file used as a reference

waveform (up to 204 characters).

It can be set in the "Browse For Folder" screen

displayed by clicking the [...] button.

• The DATATYPE folder that stores a historical data

definition file which is in the file save destination

folder of a data logging flow should be stored

together.

• Do not store a CSV file other than one used as a

reference waveform.

• Store only a CSV file and a historical data definition

file that were output in one data logging flow.

Otherwise, creating a reference waveform learning

data may fail, or created reference waveform data

may not be accurate.

• It is recommended to specify a folder other than one

used as a file save destination of the data logging

flow in operation.

• Note the following content If specifying a folder used

as a file save destination of the data logging flow.

When specifying the folder, waveform learning is

performed without checking that the saved data has

no problem. If any abnormal waveform is included in

a reference waveform, it is determined to be normal

and expected operation is not performed.

Reference Waveform*2 Set a data number to be a reference waveform (2 to

258).

The following types of data are not available for

waveform learning.

• BOOL type

• LINT type

• ULINT type

• STRING type

• WSTRING type

649 CREATING REFERENCE WAVEFORM LEARNING DATA9.1 Reference Waveform Learning Data File Creation Setting Screen

*1 Do not specify a folder with the encryption attribute enabled.*2 The data number is determined as follows based on output data of which the checkbox in the "Output" column is selected in the [Output

Data] tab in the "Data Storing Setting" screen. (N = The number counted from the top of the output data of which the checkbox in the "output" column is selected)"Data Storing Setting" screen for a data collection process: N+2"Data Storing Setting" screen for a data modification process: N+1

*3 The total number of characters in the names of an output destination folder and file name should be within 259 characters.

Reference

Waveform Learning

Data output file

path

Output folder*1,*3 Set the path of a folder in which created reference

waveform learning data is output (up to 246 characters).

It can be set in the "Browse For Folder" screen


Specify a local folder.

File Name*3 Set a file name of created reference waveform learning

data (1 to 250 characters).

Generation target

setting

Set whether to create reference waveform learning data

with this setting.

For setting only, select the "Out of generation target".

Setting of

Reference

Waveform Learning

Data creation

Learning period Specify the period for learning a reference waveform

with time starting from the first record of the reference

waveform.

Waveforms recorded after the learning period are not

included in the learning and discarded.

• Set the period so that the number of records in the

learning period is within 1000000.

• Depending on the property of a waveform used for

the creation, learning may not be performed even if

the number of records in learning period is within

1000000 (Analytical limit over error). Shorten the

learning period in this case.

• When an learning period is set long, the precision is

high, but the time for waveform learning will be long.

(Page 424 Similar Waveform Recognition Tool)

• The learning period is determined based on the date

and time of records. When the date and time of

records in the learning period is shifted back and forth

due to the daylight saving time adjustment or clock

setting change, the record in the period is increased

or decreased.

Consider the above and then set the learning period.

Unit waveform

width

Specify the range to reference an inspection waveform

with the number of records for a similarity diagnosis.

A value which is almost equivalent to the result obtained

by comparing the reference range of an inspection

waveform with a reference waveform will be the

diagnosis result.

For accurate diagnosis, set the number of records for

the unit waveform width so that a characteristic

waveform such as an impulse waveform which appears

cyclically can be included.

[OK] button Click this to apply the set contents and close the

screen.


10 PUBLISHING DATA SETTING 65

10

10 PUBLISHING DATA SETTING

This chapter shows the setting of publishing data.

Window

Click the [Publishing Data Setting] button.

Displayed items


(1) Setting list The list of publishing data settings is displayed.

One publishing data setting corresponds to one data

diagnosis flow.

Setting name The name set in the "Publishing Data Setting" screen is

displayed.

Publishing Data Setting Click this to open the "Publishing Data Setting" screen. Page 66 Publishing Data Setting

Screen




deleted.

(1)

6610 PUBLISHING DATA SETTING10.1 Publishing Data Setting Screen

10.1 Publishing Data Setting ScreenThis section shows the detailed setting of publishing data.

Window

Click the "Publishing Data Setting" column in the setting list.

Displayed items

*1 When any name is not set in "Setting name", the name of a selected publishing data definition file is set instead.*2 The value after the decimal point is rounded off.

Item Description Consideration Reference

Setting name Set the name of publishing data setting (1 to 239

characters).

Publishing data

definition

[Read file] button Click this to set a publishing data definition file.

It can be set in the "Select publishing data definition

file" screen displayed by clicking this button.*1

Set a publishing data definition file

which was output in the following

section.

Page 51 Creating a publishing

data definition file

[Publishing data

list] button

Click this to open the "Publishing data list" screen. Check that it is the same as data

which was set as the output target in

the following section.



Page 67

Publishing Data

List Screen

Diagnosis

execution cycle

setting

Diagnosis

execution cycle

Set the number of records to be an inspection

waveform needs to be input for one similarity diagnosis

The recommended value and the

upper limit value are as follows:

• Recommended value: Quarter of

the unit waveform width*2 or a value

calculated by '100 collection

interval (msec)', whichever is

smaller

• Upper limit value: Half of the unit

waveform width*2

[Apply] button Click this to apply the setting.

[OK] button Click this to close the screen after the setting is

applied.

10 PUBLISHING DATA SETTING10.2 Publishing Data List Screen 67

10

10.2 Publishing Data List ScreenThis section shows the list of publishing data.

Window

Click the [Publishing data list] button in the "Publishing Data Setting" screen.

Displayed items


Data name The name of publishing data is displayed. Check that it is the same as data

which was set as the output target in

the following section.



Data type The type of publishing data is displayed.

6811 OPERATION SETTING

11 OPERATION SETTING

This chapter shows the operation setting of similar waveform recognition.

Window

Click the [Operation Settings] button.

Displayed items


(1) Setting list The list of operation settings is displayed.

Setting name The name set in the "Operation Settings" screen is

displayed.

Operation target upon

execution

The content set for "Runtime behavior target setting" in

the "Operation Settings" screen is displayed.

Publishing Data Setting Name The name set in "Publishing Data Setting Name" in the

"Operation Settings" screen is displayed.

Operation Settings Click this to open the "Operation Settings" screen. Page 69 Operation Setting

Screen




deleted.

(1)

11 OPERATION SETTING11.1 Operation Setting Screen 69

11

11.1 Operation Setting ScreenThis section shows the detailed operation setting of similar waveform recognition.

Window

Click the "Operation Settings" column in the setting list.

Displayed items


Setting name Set the name of an operation setting (up to 32

characters).*1

• The following are displayed to avoid name

duplication by default.

Setting01, Setting02, Setting03, ...

Runtime behavior target setting Set whether to perform a similarity diagnosis

with this setting contents.

For the operation setting of a similarity

diagnosis only, select the "Out of generation

target".

Reference Waveform Learning Data File Set the path of a reference waveform learning

data file (up to 259 characters).

It can be set in the "Selection of Reference

Waveform Learning Data File" screen


• Set a file created in Similar

Waveform Recognition Tool.

I/O setting Publishing Data Setting Name Select the name of a publishing data setting

to be used among the publishing data

settings set in the following section.


If this item is changed, the contents

set in the "Diagnosis Stop Condition

Setting" screen will be blank.

Page 74 Diagnosis Stop

Condition Setting Screen

Input Data Name Select publishing data to be an inspection

waveform.

Publishing data can be selected in the

"Publishing data list" screen displayed by

clicking the [...] button.

The input data name corresponding to

the publishing data setting selected in

"Publishing Data Setting Name" is

selected.

Page 73

Publishing Data

List Screen

Response Data Name Set the name of relative similarity score that

is output as a diagnosis result and the name

of the diagnosis result (1 to 32 characters).

They are displayed by default as follows:

• Relative Similarity Score:

SimilarityScore_XX

• Diagnosis Result: Result_XX

(XX: Operation setting number (01 to 64))

For details on response data, refer to

the following:

Page 14 Similar Waveform

Recognition Tool

(3)

(1) (1)(2) (2)

(4)

7011 OPERATION SETTING11.1 Operation Setting Screen

Diagnosis

operation

setting

Diagnosis stop condition setting Set the stop condition of a diagnosis. Page 74

Diagnosis Stop

Condition

Setting Screen

Detection sensitivity setting Set the sensitivity for determining that an

inspection waveform is different from a

reference waveform with either of the

following methods:

• (1) Using the slider: Set the detection

sensitivity by using the slider.*2,*3

• (2) Entering a value directly: Set the

detection sensitivity by entering a similarity

score threshold directly.*4,*5

The slider position and the entered value link

with each other.

For the similarity score for each detection

sensitivity level, refer to the following:

Page 42 Similarity Score Threshold List

• The higher the sensitivity is set, the

more frequently a difference is

detected.

• The lower the sensitivity is set, the

higher the possibility that a

difference is missed is.

• To change the detection sensitivity

setting after starting an actual

operation, refer to the following:

Page 58 Changing the detection

sensitivity setting or diagnosis result

file saving setting

• When setting the detection

sensitivity by selecting a level

between 1 to 6, the similarity score

threshold that has six decimal

places is set according to the

selected level, not the displayed

value.

• When setting the detection

sensitivity by selecting any point

between levels with the slider or by

entering the similarity score

threshold directly, the displayed

similarity score threshold, which

has one decimal place, is set.


11 OPERATION SETTING11.1 Operation Setting Screen 71

11

Diagnosis

Result file

saving setting

Output Settings Set the output setting of a diagnosis result

file.

• Not output: A diagnosis result file is not

output.

• Only difference condition: A diagnosis

result file that includes only results

determined to be different is output.*6

• Fixed cycle output: A diagnosis result file

that includes results determined to be

similar and different is output periodically

during a diagnosis.

• Select "Only difference condition" to

display a graph of data when a

difference is detected in GX

LogViewer.

• Select "Fixed cycle output" to check

an inspection waveform and

diagnosis result in another

application.

• If selecting "Fixed cycle output," a

diagnosis result file is output during

the diagnosis; therefore, secure

free space on a disk in advance.

• To change the output setting after

starting an actual operation, refer to

the following:

Page 58 Changing the detection

sensitivity setting or diagnosis result

file saving setting

Page 87

Operational

Specifications

(3) Only

difference

condition

Numeric

representation

Set the numeric representation of numerical

values to be output in a diagnosis result file.

• Decimal representation

• Index representation

Number of

digits after

decimal point

Set the number of digits (0 to 15) after the

decimal point of a numerical value to be

output in a diagnosis result file.

• Up to 15 digits can be set when

outputting the diagnosis result file

with Similar Waveform Recognition

Tool in Real-time Data Analyzer

whose version is 1.02C or earlier.

• Up to 16 digits can be set when

outputting the diagnosis result file

with Similar Waveform Recognition

Tool in Real-time Data Analyzer

whose version is 1.03D or later.

Switched

when the

following

number of

rows (records)

is exceeded.*7

Set the number of rows of data to be stored in

one file.

A diagnosis result file is output when the set

number of rows is exceeded.

• If the number of specified rows is

small, the load on Windows will be

high because a saved file is

frequently created. In this case,

review the setting to increase the

number of rows.

Max. number

of save files

(when

exceeded, the

oldest file is

deleted.)*7

Set the upper limit number of files to be

stored in one folder.

Files are deleted in chronological order when

the set number of files is exceeded.

Save

destination

folder

Set the save destination of a diagnosis result

file (up to 213 characters).

It can also be set in the "Browse For Folder"

screen displayed by clicking the [...] button.

Refer to the following:

Page 72 Considerations for

setting a save destination folder

Prefix of file

name*7Set the prefix of a diagnosis result file name.

It is displayed by default as follows:

• LOG_XX


• Set a file name prefix that is

different from the ones output in

other applications.

Example of

save file path

The save file path is displayed according to

the current setting.


7211 OPERATION SETTING11.1 Operation Setting Screen

*1 The same name can not be used for operation settings.*2 Available only when a reference waveform learning data file is selected.*3 When a reference waveform learning data file is selected, the similarity score thresholds are displayed in "(Similarity score threshold)"

according to the file content. (If not selected, '-' is displayed.)*4 The recommended value is displayed by default.

'-' is displayed and a value cannot be entered if a reference waveform learning data file is not selected.*5 A value can be specified down to the first decimal place.*6 The result is not output while the similarity diagnosis is paused.*7 The default value is restored if the output setting is changed (to "Only difference condition" or "Fixed cycle output").

■ Considerations for setting a save destination folder • When setting a folder with restricted access as a save destination folder, check that 'Modify,' 'Read,' and 'Write' are allowed

for the account of the save destination.

• In the antivirus software settings, exclude a save destination folder from the monitoring targets. (Page 23 Exclusion

setting in antivirus software)

• When creating a file other than one to be processed (file the name of which matches the rules of the name of a file to be

processed) in a save destination folder, the operation is not guaranteed.

• Do not save a large number of files or folders to a save destination folder; otherwise, starting the operation may take time or

fail.

Diagnosis

Result file

saving setting

(4) Fixed cycle

output

Numeric

representation

Set the numeric representation of numerical

values to be output in a diagnosis result file.

• Decimal representation

• Index representation

Number of

digits after

decimal point

Set the number of digits (0 to 15) after the

decimal point of a numerical value to be

output in a diagnosis result file.

Switching

cycle setting*7Set the cycle to switch diagnosis result files.

After starting a diagnosis, recorded diagnosis

results are output in one file in the set cycle.

Max. number

of save files

(when

exceeded, the

oldest file is

deleted.)*7

Set the upper limit number of files to be

stored in one folder.

Files are deleted in chronological order when

the set number of files is exceeded.

• If the oldest file cannot be deleted,

the latest file is saved in addition to

the upper limit number of files.

Save

destination

folder

Set the save destination of a diagnosis result

file (up to 213 characters).

It can also be set in the "Browse For Folder"

screen displayed by clicking the [...] button.

Refer to the following:

Page 72 Considerations for

setting a save destination folder

Prefix of file

name*7Set the prefix of a diagnosis result file name.

It is displayed by default as follows:

• RES_XX


Set a file name prefix that is different

from the ones output in other

applications.

Example of

save file path

The save file path is displayed according to

the current setting.

[Apply] button Click this to apply the setting.

[OK] button Click this to close the screen after the setting

is applied.


11 OPERATION SETTING11.2 Publishing Data List Screen 73

11

11.2 Publishing Data List ScreenThis section shows the screen for selecting publishing data to be an inspection waveform.

Window

Click the [Publishing data list] button in the "Operation Settings" screen.

Displayed items

Item Description

Data name The name of publishing data is displayed.

Select the checkbox of publishing data to be an inspection waveform.

Data type The type of publishing data is displayed.

7411 OPERATION SETTING11.3 Diagnosis Stop Condition Setting Screen

11.3 Diagnosis Stop Condition Setting ScreenThis section shows the condition settings to stop a diagnosis.

Window

Click the [Not specified.] or [Specified.] button for the diagnosis stop condition setting in the "Operation Settings" screen.

Displayed items

*1 It will be blank and cannot be set if "Data name" is blank.*2 Available comparison operators and constant values to compare depend on the data type. For details, refer to the following:

Page 43 Pause of Similarity Diagnosis

Item Description

Condition Convergence Type Select a combination type for conditions.

• AND Convergence: A diagnosis is stopped when all the conditions are satisfied.

• OR Convergence: A diagnosis is stopped when any of the conditions is satisfied.

Data name Select publishing data to set conditions.

Data type The data type of selected publishing data is displayed.

Condition Select a comparison operator.*1,*2

Constant value Specify a constant value to compare.*1,*2

[Delete] button Click this to delete a selected row.

12 RESPONSE DATA SETTING 75

12

12 RESPONSE DATA SETTING

This chapter shows the setting of response data.

Window

Click the [Response Data Setting] button.

Displayed items

*1 Can be clicked when a checkbox in the "Output" column is selected.*2 To read an output response data definition file in Real-time Flow Designer, refer to the following:

Page 53 Reading a response data definition file*3 A file name is assigned as follows:

(Publishing data setting name)_Outputdata

Precautions

• If a response data name is changed in the "Operation Settings" screen of Similar Waveform Recognition Tool after reading

the response data definition file in Real-time Flow Designer, it is required to redo from step 5 shown in the following:


• To change response data after starting an actual operation, refer to the following:

Page 59 Changing a response data setting


(1) Setting list Output Select the checkbox of a publishing data setting to

output its response data definition file.

Publishing Data Setting Name The name of a publishing data setting is displayed. Page 65 PUBLISHING DATA

SETTING

Response Data Setting Click this to display the "Response Data List" screen. Page 76 Response Data List

Screen

[Response data definition file output] button*1,*2 Click this to output a response data definition file*3 per

publishing data setting for which the checkbox in the

"Output" column is selected.

(Response data definition files according to the number

of selected publishing data settings are output at once.)

Select the destination to save files in the "Browse For

Folder" screen that appears by clicking this button.

(1)

7612 RESPONSE DATA SETTING12.1 Response Data List Screen

12.1 Response Data List ScreenThis section shows the screen for selecting response data to be output.

Window

Click the "Response Data Setting" column in the setting list.

Displayed items

Item Description

Response Data Response data (number of different detection, relative similarity score, and diagnosis result) is

displayed.

Select the checkbox of response data to be output.

For details on response data, refer to the following:

Page 14 Similar Waveform Recognition Tool

Data Type The type of response data is displayed.

13 EXECUTION/MONITORING OF SIMILAR WAVEFORM RECOGNITION 77

13

13 EXECUTION/MONITORING OF SIMILAR WAVEFORM RECOGNITION

This section shows the operation for starting or stopping a similarity diagnosis and displaying both the execution status of

each operation setting and the waveform monitor.

Window

Click the [Execution/Monitoring] button.

(1)

(2)

(3)

7813 EXECUTION/MONITORING OF SIMILAR WAVEFORM RECOGNITION

Displayed items

*1 If the output status is "In suppression," a diagnosis is continued but a diagnosis result file is not output.*2 When stopping the operation of a similarity diagnosis after starting the operation, content displayed right before stopping remains.*3 If either of the following conditions is met, the change in the file output status is not applied to the actual operation.

- The operating status of a similarity diagnosis is not "Execution ongoing."- "Fixed cycle output" is not selected for "Output Settings" in "Diagnosis Result file saving setting."

*4 Can be used for organizing a folder. For example, a diagnosis result file can be moved to another folder by suspending the file output.*5 If the operation is suspended during a similarity diagnosis, a file is also output at that timing as well.*6 When "Execution ongoing" is displayed in 'Operation Settings', the display content in the [Monitor execution status] tab is updated every

one second.*7 Blank before starting the first similarity diagnosis.*8 "Preparing for diagnosis" is displayed immediately after the diagnosis stop condition turned from satisfied to unsatisfied.


Operation status (1) Operating status The operating status of a similarity diagnosis is displayed.

• Execution ongoing: A diagnosis is being performed.

• Stopped: A diagnosis is not performed.

• Transition to execution: Reference waveform learning data is being

read in order to perform a diagnosis.

• Transition to stop: Completion of a similarity diagnosis or diagnosis

result file saving is being waited in order to stop a diagnosis.

[Start Operation]/[Stop

Operation] button

Click this to start or stop a similarity diagnosis.

Diagnosis Result File output

status (Fixed cycle output)

(2) File output status*1,*2 The output status of a diagnosis result file is displayed. (When "Fixed

cycle output" is selected for "Output Settings")

• In suppression: Processing for outputting a diagnosis result file is

stopped.

• Canceling suppression: Processing for outputting a diagnosis

result file is in progress.

[Suppress]/[Cancel

suppression] button*3,*4Click this to switch the output operation for a diagnosis result file.

(When "Fixed cycle output" is selected for "Output Settings")

• [Suppress] button: To suspend the output of a file in a fixed cycle.*5

• [Cancel suppression] button: To resume the output of a file in a

fixed cycle.

[Monitor execution status]

tab*6Setting name*2,*7 The name set in the "Operation Settings" screen is displayed. Page 69 Operation

Setting Screen

(3) Runtime behavior target

setting*2,*7The content set for "Runtime behavior target setting" in the

"Operation Settings" screen is displayed.

Page 69 Operation

Setting Screen

Diagnostic state*2,*7,*8 The diagnosis status of a similarity diagnosis is displayed.

• Similarity: The similarity score is higher than a threshold value and

the diagnosis stop condition is not satisfied.

• Difference: The similarity score is lower than a threshold value and

the diagnosis stop condition is not satisfied.

• Preparing for diagnosis: A continuous waveform which can be

diagnosed has not been input yet and the diagnosis stop condition

is not satisfied.

• Suspended: The diagnosis stop condition is satisfied.

It will be blank when "Out of operation target" is displayed in 'Runtime

behavior target setting'.

Detailed monitoring Click this to open the "Similar Waveform Recognition execution

status detailed monitoring" screen.

It will be blank when "Stopped" is displayed in "Operation status" or

"Out of operation target" is displayed in "Runtime behavior target

setting."

Page 79 Similar

Waveform

Recognition

Execution Status

Detailed Monitoring

Screen

[Monitor waveform] tab [Start GX LogViewer] button Click this to start GX LogViewer and monitor a waveform in which a

diagnosis is being executed.

Page 90 Similar

waveform

recognition monitor

13 EXECUTION/MONITORING OF SIMILAR WAVEFORM RECOGNITION13.1 Similar Waveform Recognition Execution Status Detailed Monitoring Screen 79

13

13.1 Similar Waveform Recognition Execution Status Detailed Monitoring Screen

This section shows the details of the status of a similarity diagnosis being performed.

The display contents except for "Similarity score threshold list" are updated every one second.

Window

Click the "Detailed monitoring" column in the [Monitor execution status] tab.

Displayed items

*1 "Preparing for diagnosis" is displayed immediately after the diagnosis stop condition turned from satisfied to unsatisfied.


Setting name The name set in the "Operation Settings" screen is displayed. Page 69 Operation

Setting Screen

Diagnostic state*1 The diagnosis status of a similarity diagnosis is displayed.

• Similarity: The similarity score is higher than a threshold value and the diagnosis

stop condition is not satisfied.

• Difference: The similarity score is lower than a threshold value and the diagnosis

stop condition is not satisfied.

• Preparing for diagnosis: A continuous waveform which can be diagnosed has

not been input yet and the diagnosis stop condition is not satisfied.

• Suspended: The diagnosis stop condition is satisfied.

The latest Anomaly start date and time*2 The last date and time when "Difference" is displayed in "Diagnostic state" is

displayed.

Highest similarity score recorded date and

time*2The date and time when the highest similarity score was recorded is displayed.

Lowest similarity score recorded date and

time*2The date and time when the lowest similarity score was recorded is displayed.

Date and time of latest diagnosis*2 The date and time (local time) of the last diagnosis is displayed.

Present status continuity hours The elapsed time since the diagnosis status was changed to the current one is

displayed (hh:mm:ss).*3

Statistic information Highest similarity

score

The highest similarity score is displayed (one decimal place).*4*5

Central similarity

score

The median of the highest and lowest similarity scores is displayed (one decimal

place).*5*6

Lowest similarity

score

The lowest similarity score is displayed (one decimal place).*4*5

Difference condition

detection count

The number of times that "Difference" is displayed in "Diagnostic state" is

displayed.*7

Operating time The elapsed time since the first reference waveform was input after starting a

similarity diagnosis is displayed (hh:mm:ss)*3.

Similarity score

threshold information

Similarity score

threshold

Both the set detection sensitivity and the similarity score threshold corresponding

to the reference waveform learning data are displayed (one decimal place).

Similarity score

threshold list

The similarity score threshold corresponding to the detection sensitivity 1 to 6 is

displayed (one decimal place).*8

8013 EXECUTION/MONITORING OF SIMILAR WAVEFORM RECOGNITION13.1 Similar Waveform Recognition Execution Status Detailed Monitoring Screen

*2 It is displayed according to the format 'yyyy/MM/dd HH:mm:ss.sssssssss'. (Zero padding is applied to numbers (nine digits) from milliseconds to nanoseconds.)

*3 The upper limit of a display value is '9999:59:59'. It is displayed when a value reaches the upper limit.*4 '-' is displayed while preparing for a diagnosis (when the operation of a similarity diagnosis has just started).*5 The value is not updated while pausing a diagnosis or preparing for it (when the diagnosis has just restarted after pausing or stopping it).*6 The median calculated from the learning data is displayed while preparing for a diagnosis (when the operation of a similarity diagnosis

has just started).*7 The upper limit of a display value is '999999999'. It is displayed when a value reaches the upper limit.*8 The similarity score threshold when starting a similarity diagnosis is displayed. (The value will not be updated.)

14 OUTPUTTING DIAGNOSIS RESULT FILE 81

14

14 OUTPUTTING DIAGNOSIS RESULT FILE

A diagnosis result file records the diagnosis result of an inspection waveform in the CSV format.

Content that is output in the file differs depending on the output setting in "Diagnosis Result file saving setting" of the

"Operation Settings" screen.


When "Only difference condition" is selected for "Output Settings"A file that recorded only results determined to be different is output.

The output content can be checked in a graph by using GX LogViewer.



The following information is recorded in a diagnosis result file:

• Inspection waveform when a diagnosis was performed

• Similarity score when an inspection waveform and reference waveform learning data are compared

• A piece of reference waveform that is considered to correspond to the inspection waveform with the lowest similarity score,

in the range of an inspection waveform recorded in the file.

When "Fixed cycle output" is selected for "Output Settings"A file that recorded inspection waveforms and their diagnosis results (similar or different) is output in a fixed cycle.

An output file can be checked in another application such as Excel.

The following information is recorded in a diagnosis result file:

• Inspection waveforms to which a similarity diagnosis was being performed

• Diagnosis results (similar or different)

In Excel, a diagnosis result file can be checked in a graph as shown below. (Blue: similar, red: different)

Precautions

• To check a diagnosis result file in another application, use a copied file.

• A diagnosis result file depends on the amount of an inspection waveform that is published from Real-time Flow Manager

when performing a diagnosis. Therefore, the data volume of an output file may vary.

• For a fixed cycle output, the oldest files are deleted if the number of saved files exceeds the set upper limit. To continue

saving diagnosis result files, it is required to save (copy) the files to another folder.

Time

Inspection waveform data

8214 OUTPUTTING DIAGNOSIS RESULT FILE14.1 File Specifications

14.1 File SpecificationsThis section shows the specifications of a diagnosis result file.

*1 If the number of output files exceeds a specified number of files, files are compared by using the eight-digit sequential number of the end of a file name, and deleted in ascending order.

*2 When the file name suffix of an output file reaches 'FFFFFFFF,' take either of the following corrective actions:- Move all files in a save destination folder to another folder and click the [Start Operation] button in the "Execution/Monitoring of Similar Waveform Recognition" screen.- Set another path for the save destination folder.

*3 If the number of output files exceeds a specified number of files, files are deleted in chronological order.*4 The output content differs depending on the diagnosis status. Refer to the following:*5 It will be Shift JIS when outputting a diagnosis result file with Similar Waveform Recognition Tool in Real-time Data Analyzer whose

version is 1.02C or earlier.

Precautions

• Set a file name prefix that is different from the ones output in other applications. If it is duplicated, review the setting.

• If the number of rows specified in the following item is small, the load on Windows will be high because a saved file is

frequently created: "Switched when the following number of rows (records) is exceeded." in the "Operation Settings"

screen. In this case, review the setting to increase the number of rows.

• Do not allow other applications to access a diagnosis result file while outputting the file; otherwise, creating/updating files

may fail and cause an error.

• For considerations for a folder to save a diagnosis result file, refer to the following:

Page 72 Considerations for setting a save destination folder

Item Specification Range

File name

(When "Only

difference

condition" is

selected for

"Output Settings")

The format of a file name is 'Prefix_suffix.extension'. 45 characters

Prefix Any character(s) 32 characters

Suffix An eight-digit sequential number (hexadecimal) is added. (It is sequentially

incremented from 00000001 to FFFFFFFF.)

When saving a file with the prefix same as one saved in a save destination, the

suffix is incremented from the maximum prefix.

00000001 to

FFFFFFFF*1,*2

Extension csv

File name

(When "Fixed

output cycle" is

selected for

"Output Settings")

The format of a file name is 'Prefix_suffix.extension.'*3*4 45 characters

Prefix Any character(s) 25 characters

Suffix YYYYMMDD_hhmmss

• YYYYMMDD: Year, month, day

• hhmmss: Hour, minute, second

Extension csv

File format csv

Character encode UTF-8 (with BOM)*5

Save destination Any location

File size Not defined

14 OUTPUTTING DIAGNOSIS RESULT FILE14.2 File Structure (When "Only difference condition" is selected for "Output Settings") 83

14

14.2 File Structure (When "Only difference condition" is selected for "Output Settings")

The following figure shows the structure of a diagnosis result file that is output when "Only difference condition" is selected for

"Output Settings."

*1 When outputting a diagnosis result file with Similar Waveform Recognition Tool in Real-time Data Analyzer whose version is 1.03D or later, numbers less than a second are displayed up to nine digits with zero padding.When outputting a diagnosis result file with Similar Waveform Recognition Tool in Real-time Data Analyzer whose version is 1.02C or earlier, numbers less than a second are displayed up to three digits with zero padding.(Example) When the time stamp is 2019/1/10 3:59:00.01Version 1.03D or later: 2019/1/10 3:59:00.010000000Version 1.02C or earlier: 2019/1/10 3:59:00.010

*2 The inspection waveform and the reference waveform used for the determination when the lowest similarity score was detected are output for a number of the unit waveform width.

(1) Header

(2) Data

3SWR_2

DOUBLE[DEC.3]

DOUBLE[DEC.3]

DOUBLE[DEC.3]

DOUBLE[DEC.3]

[LOGGING]

DATETIME[YYYY/MM/DD hh:mm:ss.ns]

,

,

,

,

,

,

4 ,

,

5 CR+LF

CR+LF

SimilarityScore

DetectedAbnormal

WaveDATA1TIME , , , ,

SelectedSimilarWave

CR+LF

55.2082298.0542017/9/1

10:15:32.500000000

, , , , 2698.226 CR+LF

64.654 359.841

2298.054

54.1162698.2262017/9/1

10:15:32.600000000

, , , , 2759.424 CR+LF

2698.226

, , , , CR+LF

51.5603410.9912017/9/1

10:15:47.600000000

, , , , 3390.108 CR+LF

3410.991

52.2383267.5562017/9/1

10:15:47.700000000

, , , , 3023.256 CR+LF

3267.556

, , , , CR+LF

61.5901911.1582017/9/1

23:59:00.000000000

, , , , 1012.124 CR+LF

1911.158

359.8412017/9/1

00:00:00.000000000

, , , , 360.225 CR+LF

, , , , CR+LF

CR+LF

C:\ecop\...55.002 , , , ,1 504

, 2

(1)

(2)

Unit waveform

width

Entire inspection waveform column

Date/time column*1

Similarity score column

Inspection waveform column*2

Reference waveform column*2

End data row

Lowest similarity score

File information

row

Data type information

row

Data name

row

Start data row

Diagnosis setting

information row

8414 OUTPUTTING DIAGNOSIS RESULT FILE14.2 File Structure (When "Only difference condition" is selected for "Output Settings")

The following table shows the details of each data column.

*1 The output content differs depending on the diagnosis status. Refer to the following:Data content that is output for each diagnosis status

File information row

*1 'ECBS_1' will be output when outputting a diagnosis result file with Similar Waveform Recognition Tool in Real-time Data Analyzer whose version is 1.02C or earlier.

Diagnosis setting information row

Data type information rowThe data type corresponding to each data column is output.

However, when outputting a diagnosis result file with Similar Waveform Recognition Tool in Real-time Data Analyzer whose

version is 1.02C or earlier, the data type of the date and time column will be DATETIME [YYYY/MM/DD hh:mm:ss.sss].

Column Item Description

1st column Date and time The time stamp of a record is output.

2nd column Entire inspection waveform The value of an inspection waveform is output.

3rd column Similarity score A similarity score*1 is output.

4th column Inspection waveform The value of an inspection waveform is output.

5th column Reference waveform The value of a reference waveform is output.


1st column File type '[LOGGING]' is output.

2nd column File version 'SWR_2' is output.*1

3rd column Number for data type information

row

'3' is output.

4th column Number for data name row '4' is output.

5th column Number for data starting rows '5' is output.

6th column Number for diagnosis setting

information row

'2' is output.


1st column Similarity score threshold The similarity score threshold (six decimal places) corresponding to the detection sensitivity is output.

2nd column Detection sensitivity The set detection sensitivity is output.

3rd column Diagnosis execution cycle The set diagnosis execution cycle is output.

4th column Unit waveform width The unit waveform width of reference waveform learning data is output.

5th column Reference waveform learning data

path

The file path of reference waveform learning data is output.

14 OUTPUTTING DIAGNOSIS RESULT FILE14.3 File Structure (When "Fixed output cycle" is selected for "Output Settings") 85

14

14.3 File Structure (When "Fixed output cycle" is selected for "Output Settings")

The following figure shows the structure of a diagnosis result file that is output when "Fixed cycle output" is selected for

"Output Settings."

*1 The value less than a second is displayed with zero padding up to the 9th digit.

(1) Header

(2) Data

3SWR_2

DOUBLE[DEC.3]

DOUBLE[DEC.3]

DOUBLE[DEC.3]

DOUBLE[DEC.3] DIAGRESULT

[LOGGING]

DATETIME[YYYY/MM/DD hh:mm:ss.ns]

,

,

,

,

,

,

4 ,

,

5 CR+LF

CR+LF

SimilarityScore

DetectedAbnormal

WaveDATA1TIME , , , ,

SelectedSimilarWave

DiagnosisResult

CR+LF

55.2082298.0542017/9/1

10:15:32.500000000

, , , , 2698.226 CR+LF

100 359.841

2298.054

54.1162698.2262017/9/1

10:15:32.600000000

, , , , 2759.424 CR+LF

2698.226

, , , , CR+LF

51.5603410.9912017/9/1

10:15:47.600000000

, , , , 3390.108 CR+LF

3410.991

52.2383267.5562017/9/1

10:15:47.700000000

, , , , 3023.256 CR+LF

3267.556

, , , , CR+LF

61.5901911.1582017/9/1

23:59:00.000000000

, , , , 1012.124 CR+LF

1911.158

359.8412017/9/1

00:00:00.000000000

, , , , 360.225 CR+LF

, , ,

0

1

1

1

0

0

, CR+LF

CR+LF

C:\ecop\...55.002 , , , ,1 504

, 2

(1)

(2)

Entire inspection waveform column

Date/time column*1

Similarity score column

Inspection waveform column

Reference waveform column

Diagnosis result column

End data row

File information

row

Data type information

row

Data name

row

Start data row

Diagnosis setting

information row

8614 OUTPUTTING DIAGNOSIS RESULT FILE14.3 File Structure (When "Fixed output cycle" is selected for "Output Settings")

The following table shows the details of each data column.

*1 The output content differs depending on the diagnosis status. Refer to the following:Data content that is output for each diagnosis status

• Data content that is output for each diagnosis status

The following shows the data content that is output for each status of a similarity diagnosis.

*1 A diagnosis file is not output if the diagnosis status is "Suspended."

File information row

Diagnosis setting information row

Data type information rowThe data type corresponding to each data column is output.


1st column Date and time The time stamp of a record is output.

2nd column Entire inspection waveform The value of an inspection waveform is output.

3rd column Similarity score A similarity score*1 is output.

4th column Inspection waveform The value of an inspection waveform is output.

5th column Reference waveform The value of a reference waveform is output.

6th column Diagnosis result The diagnosis result ('0' or '1')*1 is output.

• 0: Similar

• 1: Different

Item Output content for each diagnosis status

Preparing for diagnosis

Similarity Difference Suspended

Similarity score 100 Not defined Not defined *1

Diagnosis result 0 0 1 *1


1st column File type '[LOGGING]' is output.

2nd column File version 'SWR_2' is output.

3rd column Number for data type information

row

'3' is output.

4th column Number for data name row '4' is output.

5th column Number for data starting rows '5' is output.

6th column Number for diagnosis setting

information row

'2' is output.


1st column Similarity score threshold The similarity score threshold (six decimal places) corresponding to the detection sensitivity is output.

2nd column Detection sensitivity The set detection sensitivity is output.

3rd column Diagnosis execution cycle The set diagnosis execution cycle is output.

4th column Unit waveform width The unit waveform width of reference waveform learning data is output.

5th column Reference waveform learning data

path

The file path of reference waveform learning data is output.

14 OUTPUTTING DIAGNOSIS RESULT FILE14.4 Operational Specifications 87

14

14.4 Operational SpecificationsThis section shows the operational specifications for outputting a diagnosis result file.

A diagnosis result file is output when a diagnosis is performed or when any of the following conditions are satisfied.

When "Only difference condition" is set as a condition to output a diagnosis result fileData within the different state period is output.

A file is output at the timing when the different state period ends.

Whether the different state period has ended or not is determined as follows:

• The similar state is detected ten times in a row.

In the similar state period, if the row in which a diagnosis result file is switched is reached, the result so far will be output to the

current file and the subsequent result will be recorded in the next file.

The file to be output when the number of file switching row is '5' is described below:

Output condition Reference

Only difference condition Page 87 When "Only difference condition" is set as a condition to

output a diagnosis result file

Fixed cycle output Page 88 When "Fixed cycle output" is set as a condition to output

a diagnosis result file

: Similar state

: Different state

(1) Different state period

(a) A file is output because the similar state was detected ten times in a row.

: Similar state

: Different state

(1) Different state period 1



(a) A file is output as the file switching row (5) is reached.

(b) A file is output as the similar state was detected ten times in a row.

(1)

(a)

Time

(1)

(b)(a) (a)

(2) (3)

Time

8814 OUTPUTTING DIAGNOSIS RESULT FILE14.4 Operational Specifications

When "Fixed cycle output" is set as a condition to output a diagnosis result fileData within the execution period of a similarity diagnosis is output in a file.

Once the operation of a similarity diagnosis is started and the diagnosis is performed (an inspection waveform is determined

to be similar or different), data is accumulated as a diagnosis result file, and the inspection waveform and its diagnosis result

are output in the CSV file format according to the cycle set in "Diagnosis Result file saving setting" of the "Operation Settings"

screen.

The timings to output a file are as follows:

If a diagnosis is stopped or paused, a file is output at that time.

• A fixed cycle output is started after the diagnosis status turns from "Preparing for diagnosis" to "Similarity" or

"Difference."

• Outputting a diagnosis file during a diagnosis can be suspended. (Page 77 EXECUTION/

MONITORING OF SIMILAR WAVEFORM RECOGNITION)

• While referring to a diagnosis result file, the oldest file may not be deleted even if the set maximum number

of saved files is reached. However, the latest file is stored even if the oldest file cannot be deleted. (In this

case, the number of saved files will be 'set maximum number of saved files + 1.')

Precautions

• Data is not accumulated as a diagnosis result file while pausing a diagnosis or suspending the file output.

• If the file output is resumed while preparing for a diagnosis, data accumulated while preparing for the diagnosis is output.

However, it is output only when any data is accumulated.

• When outputting data accumulated while preparing for a diagnosis, the similarity score is output as 100 since the diagnosis

is not performed.

• If a diagnosis result file cannot be saved to a save destination folder, up to 10 files are accumulated and they will be saved

when the save destination folder becomes available. However, if there are more than 10 files to be accumulated, files are

deleted in chronological order.

• The time stamp of an output file may not be consistent with the cycle set in "Switching cycle setting" of "Diagnosis Result

file saving setting."

: Similar state

: Different state

(a) A file is output in the set cycle.

: Similar state

: Different state

(1) A diagnosis is stopped or paused. (A file is not output while stopping or pausing a diagnosis.)

(a) A file is output in the set cycle.

(b) A file is output as a diagnosis is stopped or paused.

(a) (a) (a) (a) (a)

Time

(a)(a) (a) (a) (b)(a)

(1)

Time

14 OUTPUTTING DIAGNOSIS RESULT FILE14.4 Operational Specifications 89

14

• A reference waveform is determined when a diagnosis is performed.

(Example) Timing to determine a reference waveform when the unit waveform width is five records and the diagnosis

execution cycle is three records

If a diagnosis result file is output before a reference waveform is determined, an inspection waveform is output for the content

of reference waveform (SelectedSimilarWave).

(Example) Case in which an inspection waveform is output for the content of reference waveform (SelectedSimilarWave)

when the unit waveform width is five records and the diagnosis execution cycle is three records

(1) An inspection waveform for the unit waveform width is accumulated.

(2) A diagnosis is performed as an inspection waveform for the unit waveform width is accumulated. (A reference waveform that is the most similar to the

inspection waveform No.1 to 5 is determined.)

(3) An inspection waveform is accumulated as it is not in the diagnosis execution cycle.

(4) A diagnosis is performed as it is in the diagnosis execution cycle. (A reference waveform that is the most similar to the inspection waveform No.4 to 8 is

determined.)

: Unit to determine a reference waveform (for (2))

: Unit to determine a reference waveform (for (4))

(1) An inspection waveform for the unit waveform width is accumulated.

(2) A diagnosis is performed as an inspection waveform for the unit waveform width is accumulated. (A reference waveform that is the most similar to the

inspection waveform No.1 to 5 is output.)

(3) An inspection waveform is accumulated as it is not in the diagnosis execution cycle.

(4) An inspection waveform is accumulated since it is not in the diagnosis execution cycle. (The inspection waveform is output for the content of reference

waveform (SelectedSimilarWave) since a diagnosis is not performed.)

: A reference waveform is output.

: An inspection waveform is output.

1 01:00:00 -61.925 1 0

2 01:00:01 -65.113 2 0

3 01:00:02 -59.335 3 0

4 01:00:03 -38.017 4 0

5 01:00:04 -29.051 5 0

6 01:00:05 -10.721 6 1

7 01:00:06 -2.951 7 2

8 01:00:07 3.824 8 3

No.

(1)

(1)

(1)

(1)

(3)

(3)

(4)

(2)

Time stamp Inspection waveformNumber of accumulated inspection waveforms

Diagnosis execution cycle

1 01:00:00 -61.925 1 0

2 01:00:01 -65.113 2 0

3 01:00:02 -59.335 3 0

4 01:00:03 -38.017 4 0

5 01:00:04 -29.051 5 0

6 01:00:05 -10.721 6 1

7 01:00:06 -2.951 7 2

No.

(1)

(1)

(1)

(1)

(3)

(4)

(2)

Time stamp Inspection waveform Diagnosis execution cycle

Number of accumulated inspection waveforms

9015 GX LogViewer COOPERATION15.1 Similar waveform recognition monitor

15 GX LogViewer COOPERATION

The following functions can be performed by using Similar Waveform Recognition Tool and GX LogViewer together.

For GX LogViewer, refer to the following manual.

GX LogViewer Version 1 Operating Manual

GX LogViewer Version 1 Operating Manual (Edgecross)

15.1 Similar waveform recognition monitorBy connecting GX LogViewer with Similar Waveform Recognition Tool, the following data can be checked: an inspection

waveform which is input for a similarity diagnosis and the variations of a similarity score acquired as a diagnosis result.

Displayed items

Considerations

■ Number of simultaneously-displayed diagnoses • Number of simultaneously-displayed diagnoses is two in maximum.

■ Operating status of a similarity diagnosis • A similarity diagnosis to be monitored must be in operation. If the diagnosis is stopped, the similar waveform recognition

monitor function is also stopped.

• When a similarity diagnosis is performed again, it is necessary to perform the similar waveform recognition monitor function

again.


Similar waveform recognition monitor To check the following data with a graph in real time: an inspection waveform

which is input for a similarity diagnosis and the variations of a similarity score

acquired as a diagnosis result.

Page 90 Similar waveform

recognition monitor

Diagnosis result file display To check the content of a diagnosis result file by graphing it. Page 91 Diagnosis result file

display

Item Description

(1) Inspection waveform monitor An inspection waveform to be input is displayed in real time.

When a diagnosis result is determined to be different, an inspection waveform is displayed in a different color from the

one for when the result is determined to be similar.

While a diagnosis is paused, an inspection waveform is displayed in a same color as the one for when the result is

determined to be similar.

(2) Similarity score monitor A similarity score to be a diagnosis result is displayed in real time.

While a diagnosis is paused, the similarity score is 100.

(3) Different waveform A waveform diagnosed as different from the reference waveform is displayed for each unit waveform width.

Up to four different waveforms can be displayed and waveforms are deleted in chronological order.

(4) Reference waveform The snapshots of the reference waveform most similar to a different waveform being displayed are displayed for each

unit waveform width.

(1)

(2)

(3)

(4)

15 GX LogViewer COOPERATION15.2 Diagnosis result file display 91

15

15.2 Diagnosis result file displayThe contents of a diagnosis result file which was output by Similar Waveform Recognition Tool can be displayed with the

contents similar to those displayed when the similar waveform recognition monitor function was performed.

Displayed items

Precautions

• An error occurs when displaying a diagnosis result file output with Similar Waveform Recognition Tool in Real-time Data

Analyzer whose version is 1.03D or later in GX LogViewer whose version is 1.88S or earlier. In this case, display the file

after updating GX LogViewer to version 1.90U or later.

• While a similarity diagnosis is paused, a diagnosis result file is not output.

• An error occurs when reading an output diagnosis result file if "Fixed cycle output" is selected for "Output Settings" in

"Diagnosis Result file saving setting" of the "Operation Settings" screen.

Item Description

(1) Inspection waveform An inspection waveform included in a diagnosis result file is displayed.

(2) Similarity score A similarity score included in a diagnosis result file is displayed.

(3) Different waveform An inspection waveform which is considered to be a different waveform with the lowest similarity score in a diagnosis

result file is displayed for each unit waveform width.

(4) Reference waveform The reference waveform most similar to a different waveform being displayed is displayed for each unit waveform

width.

(1)

(2)

(3)

(4)

9216 CONSIDERATIONS FOR SIMILAR WAVEFORM RECOGNITION16.1 Waveforms Used for Similar Waveform Recognition

16 CONSIDERATIONS FOR SIMILAR WAVEFORM RECOGNITION

The following shows the considerations for similar waveform recognition.

16.1 Waveforms Used for Similar Waveform Recognition

Similar waveform recognition is a function to learn a normal waveform and detect an error when a waveform different from the

normal one is input. It is suited to determine a cyclic waveform in which the same pattern of waveforms are repeated at

regular interval.

The cycle of a cyclic waveform is not limited, but parameters must be set according to the cycle. ( Input

Waveform (Reference Waveform) and Setting Items in Waveform Learning)

Observe the following to ensure more accurate diagnosis result in a similar waveform recognition.

• Specify an appropriate unit waveform width to include a specific waveform. ( Page 94 Input Waveform (Reference

Waveform) and Setting Items in Waveform Learning)

• The unit waveform width has the upper limit, so make the collection cycle longer or modify data so that all records*1 in a

specific waveform is included in the unit waveform width.

*1 Data per row is referred to as a 'record'.

• Use data with resolution which allows for detecting a difference in a specific waveform. (It is recommended data with 500

steps or more of resolution from minimum value to maximum value.)

If the resolution is too low, an expected diagnosis result may not be acquired.

(Example) To detect the level difference '0.5' of a waveform of which the minimum value is '0.0' and the maximum value is

'1000.0', resolution '0.5' or higher is necessary.

• The input waveform value should be within the range from '2.11996057443428e+152' to '-2.11996057443428e+152'.

Note that the range suitable for similar waveform recognition is '9.00719925474099e+12' to '-9.00719925474099e+12'.

16 CONSIDERATIONS FOR SIMILAR WAVEFORM RECOGNITION16.1 Waveforms Used for Similar Waveform Recognition 93

16

Properties of a waveform suitable for similar waveform recognition

Properties of a waveform not suitable for similar waveform recognition

Waveform property Example Remarks

Cyclic and amplitude is not too small.

Multiple patterns of correct

phenomena are recorded.

The interval is not required to be even.

While two or more phenomena can be

diagnosed, too many phenomena increases the

possibility of incorrect determination.

The unit waveform width is not too

short for one phenomenon.

(1) Too short and may be confused with other portions.

(2) Suitable because it includes a feature for distinction from other

portions.

If a single phenomenon is too long to be included

in the unit waveform width, shorten the collection

cycle.

Waveform property Example Remarks

Not cyclic

Waveforms with different cycles are

overlapped (noise or others).

While this example shows a case where a short-

cycle waveform is overlapped, a long-cycle

waveform also decreases the determination

accuracy.

Attribute is too small for resolution. It is recommended data with 500 steps or more of

resolution from minimum value to maximum

value.

(1) (1) (2)

9416 CONSIDERATIONS FOR SIMILAR WAVEFORM RECOGNITION16.2 Input Waveform (Reference Waveform) and Setting Items in Waveform Learning

16.2 Input Waveform (Reference Waveform) and Setting Items in Waveform Learning

• For accurate learning and waveform monitoring, set the number of records for the unit waveform width so that a

characteristic waveform such as an impulse waveform which appears cyclically can be included. However, it cannot be set

exceeding the learning period.

• For learning waveforms, input only waveforms in the similar state. If a waveform with an error is input or some noise is

included on a waveform, the determination accuracy may be decreased.

If a fluctuation or deviation is large in waveforms in the similar state, more accurate reference waveform learning data can

be generated by extending the learning period.

• When the collected data is not enough for the learning period, learning will end at the end of the collected data.

In addition, learning will also end when the collected data is enough for the learning period and the learning period is longer

than the period equivalent to 1000000 records.

• The number of records of a reference waveform necessary for waveform learning must be four times or more of the unit

waveform width within a section in which dates and times are arranged at regular intervals in chronological order. However,

even if the above condition is satisfied, sufficient diagnostic accuracy is not guaranteed. To increase the accuracy, extend

the learning period so that the number of records of reference waveform is longer than four times of unit waveform width.

16.3 Setting Items for a Data Diagnostic Flow • Set the same collection cycle to the data analysis phase and data diagnosis phase. Otherwise, the determination will be

inaccurate.

• If the collection cycle is shorter than the processing period, waveform monitoring processing cannot complete and a

process data buffer overflow error occurs. In this case, lengthen the collection cycle or the data diagnosis period. (When

changing the collection cycle, it is required to redo from the data analysis phase.)

• If the data diagnosis cycle is too long comparing the unit waveform width, waveform monitoring becomes sparse and a gap

is made in the monitoring. Therefore, be sure to set a value for data diagnosis cycle which is appropriate for the unit

waveform width. (Half of the unit waveform width is a guide)

*1 Number of contact points (records).

(1) Data diagnosis period = 5

(2) Unit waveform width*1 = 5

(3) Waveform monitoring unperformed period

(2)

(1)

(3)

16 CONSIDERATIONS FOR SIMILAR WAVEFORM RECOGNITION16.4 Restrictions on Reference Waveform Learning Data Once Created 95

16

16.4 Restrictions on Reference Waveform Learning Data Once Created

• Reference waveform learning data once created cannot be changed.

• The reference waveform learning data created once cannot be enhanced by making it learning again.

16.5 Number of Times of DetectionIn similar waveform recognition, differences are output multiple times successively for one continuous difference. In order not

to feed back for every time, make an inhibition period for a feedback execution.

Ex.

Case in which the detection sensitivity is 3:

: Determined to be similar

: Determined to be different

(1) Similarity score



5

6

4

3(2)

(1)

(3)

2

1

76

82

71

66

60

55

9616 CONSIDERATIONS FOR SIMILAR WAVEFORM RECOGNITION16.6 Relativity between Collection Data Transmission Delay and a Diagnosis Result

16.6 Relativity between Collection Data Transmission Delay and a Diagnosis Result

When collecting data by using a data collector in which the transmission delay occurs, the values of an inspection waveform

may be fluctuated because of inconstant transmission delay time. If it occurs, a waveform with a high similarity may be

determined to be different, and vice versa. If an expected diagnosis result is not obtained by such a data collector, use a

different data collector of which the fluctuation of transmission delay time is not large or shorten the collection interval.

The relativity between the similarity score and the fluctuation of a waveform included in a reference waveform is as follows:

For the transmission delay time, refer to the manual of the data collector to use.

Example of a diagnosis in which a similarity score is determined lower than expectedA reference waveform similar to an inspection waveform after fluctuation is not found, so it is determined to be different.

Reference waveform group status

Similarity score status

Little fluctuation • Similarity score tends to be high.

• Slight misalignment of waveform greatly affects the similarity score.

Lot of fluctuation • Similarity score tends to be low.

• Slight misalignment of waveform does not greatly affect the similarity score.



(3) Waveforms are compared.

(4) Fluctuation is generated.

(1)

(2)

(3)

(4)

16 CONSIDERATIONS FOR SIMILAR WAVEFORM RECOGNITION16.6 Relativity between Collection Data Transmission Delay and a Diagnosis Result 97

16

Example of a diagnosis in which a similarity score is determined higher than expectedThe similarity is increased with the learning of the reference waveform similar to an inspection waveform after fluctuation, so it

is determined to be similar.



(3) Waveforms are compared.

(4) Partially fluctuates

(5) Abnormal waveform to detect

(1)

(2)

(3)

(5)

(4)

9817 TROUBLESHOOTING BY SYMPTOM


If a function of Similar Waveform Recognition Tool does not perform properly, check the applicable items in the following and

troubleshoot the error.

Symptom Check item Corrective action

Failed to learn a waveform. Is the number of records of the

reference waveform equal to or

greater than a value of the unit

waveform width 4?

• Increase the recording time of the reference waveform, and use a reference

waveform equal to or greater than a value of the unit waveform width 4.

(The minimum required number of records is a value of the unit waveform

width 4. Use a reference waveform of longer time to increase accuracy.)



Failed to learn a waveform using a

reference waveform collected by

Real-time Flow Manager due to a

time stamp error.

Is the number of digits of the reserved

word included in the time stamp

format coincident with the number of

digits of the time stamp?

• Re-collect a reference waveform by using Real-time Flow Manager.

Failed to read a publishing data

definition file.

Does the version of Similar

Waveform Recognition Tool support

the one of Edgecross Basic Software

in which the publishing data definition

file was created?

Re-read the publishing data definition file that was created in Edgecross Basic

Software supported by the version of Similar Waveform Recognition Tool.


17 TROUBLESHOOTING BY SYMPTOM 99

17A waveform that should not be

detected may be detected incorrectly.

Can a change of a waveform of input

data be detected?

• Decrease the collection interval to detect a change of a waveform of input

data.

• Increase the unit waveform width to include a characteristic waveform.


Is an input waveform with a same

condition as when recording a

reference waveform used?

• Check the setting of Real-time Flow Designer, and use a same setting as

when learning a waveform.

• Check whether the system environment has a same condition as when

recording a reference waveform, and match the condition.



Are noise or others in processing

target input data superimposed?

• Decrease the sensitivity for detection in order not to incorrectly detect a

waveform that should not be detected.

• Take measures against noise and congestion.

• Increase the accuracy by increasing the amount of a reference waveform in

waveform learning.

• Use the data modification function to remove the noise.





Are ripples with long cycles or others

superimposed on processing target

input data?



• Take measures against noise and congestion.

• Increase the accuracy by increasing the amount of a reference waveform to

be equal to or greater than four cycles of a long cycle waveform in

waveform learning.

• Use the data modification function to remove the noise.




Page 56 Data diagnosis cycle adjustment

Is data the cycle of which fluctuates a

processing target?



• Decrease the unit waveform width in the range in which a characteristic

waveform can be included.

• Increase the accuracy by increasing the amount of a reference waveform in

waveform learning.



Is a reference waveform learning

data file created by a processing

target used?

• Specify a correct reference waveform learning data file.

• Record the processing target, and use the data as a reference waveform to

create data diagnosis parameters.


Is a data collector the transmission

delay time of which fluctuates widely

used?

• Use a data collector the transmission delay time of which fluctuates

narrowly.

• Decrease the collection interval less than the interval at which a value of

data to be collected is updated.


Page 96 Relativity between Collection Data Transmission Delay and a

Diagnosis Result

Does a reference waveform included

in the analysis period exceed

1000000 records?

• Reduce the number of records included in the analysis period to 1000000 or

less.


Is a similarity score low even when it

is not applicable to the above

patterns?

• Extend the recording time of a reference waveform to be entered when

learning a waveform, and increase the analysis period.



Page 41 Similarity Score




Is the sensitivity for detection too

high?

• If setting the detection sensitivity by adjusting the level, lower the level.

• If setting the detection sensitivity by entering a value directly, set a lower

value.



A waveform that should be detected

may be missed.

Is the unit waveform width too short? • Increase the unit waveform width to include a characteristic waveform.


Is the sensitivity for detection high

enough?

• If setting the detection sensitivity by adjusting the level, increase the level.

• If setting the detection sensitivity by entering a value directly, set a higher

value.


data be detected?


data.













Is data missing in the reference

waveform?

• Use a reference waveform in which no data is missing to learn a waveform.



Has data missing occurred in a data

collector?

• Increase the cycle for performing a data diagnosis.

• Decrease the unit waveform width.

• Terminate any unnecessary applications.

• Reduce the number of similarity diagnosis settings.

• Increase the collection interval.

• Reduce the number of data logging flows or data diagnosis flows of Real-

time Flow Manager.

• Reduce the load of other applications.

Is an error included in a reference

waveform to be entered when

learning a waveform?

• Use data in which an error is not included, and create reference waveform

learning data.


Is the length of a reference waveform

to be entered when learning a

waveform sufficient?

• Extend the recording time of a reference waveform to be entered when

learning a waveform, and increase the analysis period.



Is the resolution of the level of a

reference waveform sufficient?

• Decrease the collection cycle so that the resolution will be 500 or more.


• Use an input waveform with a resolution higher than a level difference that

should be detected.

(When detecting the level difference of '0.5', the resolution must be equal to or

less than '0.5'.)





Is a data collector the transmission

delay time of which fluctuates widely

used?

• Use a data collector the transmission delay time of which fluctuates

narrowly.

• Decrease the collection interval less than the interval at which a value of

data to be collected is updated.


Page 96 Relativity between Collection Data Transmission Delay and a

Diagnosis Result

Is the cycle for performing a data

diagnosis equal to or less than the

one-half of a unit waveform width?

• Decrease the cycle for performing a data diagnosis to equal to or less than

the one-half of the unit waveform width.

Page 56 Data diagnosis cycle adjustment

Does a reference waveform included

in the analysis period exceed

1000000 records?

• Reduce the number of records included in the analysis period to 1000000 or

less.


Has a communication error occurred

between Real-time Flow Manager

and an MQTT broker when

cooperating with Real-time Flow

Manager that is installed in another

industrial PC?

• Check the communication cable status.

• Check if there is any problem in communication settings or route between

Real-time Flow Manager and the MQTT broker.


• Reduce the number of units of data output from Real-time Flow Manager.



Is the industrial PC in which

Edgecross Basic Software is installed

running?

Run the industrial PC in which Edgecross Basic Software is installed.



17The status will be in the state in which

the different state is always detected.

Is the sensitivity for detection too

high?

• If setting the detection sensitivity by adjusting the level, lower the level.

• If setting the detection sensitivity by entering a value directly, set a lower

value.

Is a reference waveform learning

data file created by a processing

target used?

• Specify a correct data reference waveform learning data file.

• Record the processing target, and use the data as a reference waveform to

create a reference waveform learning data file.



data be detected?


data.


















industrial PC?










running?


The status will be in the state in which

the similar state is always detected.

Is a value of processing target input

data changed?

• Check the connection with a device, collection settings, or others.



data be detected?


data.









industrial PC?










running?


Is the detection sensitivity too low? • If setting the detection sensitivity by adjusting the level, increase the level.

• If setting the detection sensitivity by entering a value directly, set a higher

value.



The load on an industrial PC is high. Is the load on an industrial PC high? • Increase the cycle for performing a data diagnosis.

• Decrease the unit waveform width.

• Terminate any unnecessary applications.

• Reduce the number of similarity diagnosis settings.


• Reduce the number of data logging flows or data diagnosis flows of Real-

time Flow Manager.

• Reduce the load of other applications.

• Change the environment settings for Windows.

Page 440 Environment Settings for an Industrial PC where Real-time

Data Analyzer Runs

• Set the following items so that they will be executed when Real-time Data

Analyzer is not running:

Windows Update

Optimize Drives

Disk Cleanup

• If selecting "Only difference condition" for "Output Settings" in "Diagnosis

Result file saving setting," increase the number of rows for switching files.

• If selecting "Fixed cycle output" for "Output Settings" in "Diagnosis Result

file saving setting," increase the cycle for switching files.

A similarity diagnosis stopped. Did a diagnosis stop because the

limit of processing performance was

exceeded?

"Error in the number of connections"

is displayed when opening the

operation setting while performing the

similar waveform recognition monitor

function of GX LogViewer.

Is a same operation setting as the

opened one to be monitored in a tool

other than GX LogViewer being

operated?

End monitoring of the same operation setting in another GX LogViewer.

The following message appears

when loading a diagnosis result file in

GX LogViewer.

'The Data Logging file cannot be

displayed because the file version is

too new or the file is not

corresponding to the tool.'

Is the version of GX LogViewer 1.90U

or later?

Please contact your local Mitsubishi representative and acquire the latest GX

LogViewer, then install it.

"Preparing for diagnosis" is displayed

as the operating status in the

"Execution/Monitoring of Similar

Waveform Recognition" screen even

though a similarity diagnosis is

started.

Is a publishing data definition file

which was output from a data

diagnosis flow different from one to

be diagnosed loaded?

Reset the publishing data setting and load the publishing data definition file

which was output from the data diagnosis flow to be diagnosed.

Is an operation of Real-time Flow

Manager started?

Start the operation of Real-time Flow Manager.

Did an error occur in Real-time Flow

Manager?

Clear the error in Real-time Flow Manager.






industrial PC?










running?


"A communication message of

unsupported version was received."

is displayed during a similarity

diagnosis.

Does the version of Similar

Waveform Recognition Tool support

the one of Edgecross Basic

Software?

Use Edgecross Basic Software which is supported by the version of Similar

Waveform Recognition Tool.


"An invalid communication message

was received." is displayed during a

similarity diagnosis.

Is there any difference in the contents

of the edge application diagnosis

(MQTT) setting of Real-time Flow

Designer and the publishing data

setting of Similar Waveform

Recognition Tool?

Output the publishing data definition file again in the edge application

diagnosis (MQTT) setting of Real-time Flow Designer and reset the publishing

data setting of Similar Waveform Recognition Tool.



17Contents in the screen are not

displayed properly.

Is a value other than 100% set for

"Change the size of text, apps, and

other items" in Windows?

• Change the value to 100% for "Change the size of text, apps, and other

items".

• For Windows 10 (version 1703 or later)*1, the display of a screen can be

enlarged with high DPI scaling by using a Windows 10 function*2.

Select the execution file*3,*4, then select [Properties] on the right-click

menu.

Select the checkbox of "Override high DPI scaling behavior. Scaling

performed by:" in the [Compatibility] tab, then select "System" from the pull-

down list.

Click the [OK] button.

The same error occurs even after the

corrective action was taken.

Is antivirus software installed? Refer to the following section and configure the setting to exclude folders from

monitoring targets of antivirus software.

Page 23 Exclusion setting in antivirus software

The diagnosis status does not turn to

the 'suspended' state.

Is content of publishing data

consistent with that of a diagnosis

stop condition?

Review the settings of the diagnosis stop condition.






industrial PC?










running?


The diagnosis status does not turn to

the 'preparing for diagnosis' state

from the 'suspended' state.

Is content of publishing data

consistent with that of a diagnosis

stop condition?

Review the settings of the diagnosis stop condition.






industrial PC?










running?


Unable to display an inspection

waveform and its diagnosis result in a

graph in another application.

Are an inspection waveform and its

diagnosis result (similar or different)

displayed properly in a graph?

• Check if "Fixed cycle output" is selected for "Output Settings" in "Diagnosis

Result file saving setting."

• Check if the file to be displayed in a graph is a diagnosis result file.

• Check if a cycle for switching diagnosis result files is long.Can the diagnosis result file to be

displayed in a graph be read?

Does the diagnosis result file to be

displayed in a graph exist?

A diagnosis result file is not stored

even when "Canceling suppression"

is displayed in "Diagnosis Result File

output status (Fixed cycle output)" of

the "Execution/Monitoring of Similar

Waveform Recognition" screen.

Is "Only difference condition"

selected for "Output Settings" in

"Diagnosis Result file saving

setting"?

• Check if a large value is set for the number of rows for switching diagnosis

result files.

• Check if the save destination folder for the diagnosis result file is correct.

• Check if the operating status of Real-time Flow Manager is STOP or if an

error occurs in Real-time Flow Manager.

Is "Fixed cycle output" selected for

"Output Settings" in "Diagnosis

Result file saving setting"?

• Check if a cycle for switching diagnosis result files is long.


• Check if the operating status of Real-time Flow Manager is STOP or if an

error occurs in Real-time Flow Manager.

Is "Not output" selected for "Output

Settings" in "Diagnosis Result file

saving setting"?

Change the output setting for a diagnosis result file.

Is there enough free space in a disk

specified as the save destination?

Secure free space in the disk.

Can a diagnosis result file be

updated?

• Check if the diagnosis result file is being accessed.




*1 The Windows version can be checked by the following procedure. Press Windows key + , or select [Windows System] [Run] from Windows Start. Enter 'winver' in the "Run" screen. Check the version in the displayed screen.

*2 The display will be blurred by enlarging.The following lists the setting values for "Change the size of text, apps, and other items" and the recommended display resolution for each setting value in Windows 10.Setting value: 100%, display resolution: 1280 800 dots or moreSetting value: 125%, display resolution: 2048 1536 dots or moreSetting value: 150%, display resolution: 2048 1536 dots or moreSetting value: 175%, display resolution: 2880 1620 dots or moreSetting value: 200%, display resolution: 2880 1620 dots or moreSetting value: 225%, display resolution: 3840 2160 dots or moreSetting value: 250%, display resolution: 3840 2160 dots or more

*3 The execution file is stored in the folder in which Real-Time Data Analyzer is installed.The following are examples of storage locations.(Example) When the installation destination folder is set as the default:Screen for creating reference waveform learning data: C:\Mitsubishi Electric Edgecross Software\RDA\RWR\StandardWaveDataCreation.exe (Page 62 CREATING REFERENCE WAVEFORM LEARNING DATA)Home screen, a screen for the publishing data setting, and a screen for the operation setting: C:\Mitsubishi Electric Edgecross Software\RDA\RWR\SimilarWaveRecognitionConfig.exe(Page 33 Home screen, Page 65 PUBLISHING DATA SETTING, Page 68 OPERATION SETTING)Screen for executing or monitoring the similar waveform recognition: C:\Mitsubishi Electric Edgecross Software\RDA\RWR\SimilarWaveRecognitionMonitor.exe (Page 77 EXECUTION/MONITORING OF SIMILAR WAVEFORM RECOGNITION)

*4 The following are examples of storage locations for the execution file selected when enlarging the display of a screen of License Manager and the utility for switching language.(Example) When the installation destination folder is set as the default:License Manager: C:\Mitsubishi Electric Edgecross Software\License Manager\License Manager.exeUtility for switching language: C:\Mitsubishi Electric Edgecross Software\RDA\LanguageSwitchingUtility\RDA_LanguageSwitchingUtility.exe

*5 It is stored under the folder specified when installing Real-time Data Analyzer.The following is an example of a storage location.(Example) When the installation destination folder is set as the default:

File output does not stop even if "In

suppression" is displayed in

"Diagnosis Result File output status

(Fixed cycle output)" of the

"Execution/Monitoring of Similar

Waveform Recognition" screen.

Is "Fixed cycle output" selected for

"Output Settings" in "Diagnosis

Result file saving setting"?

Change the output setting for a diagnosis result file.

Unable to select a diagnosis result as

an execution condition for feedback

in Real-time Flow Designer.

Is an item that is desired to set as a

feedback condition displayed as

output data after acquiring the

definition information of response

data?

• Check if the definition information of selected response data is correct.

• Check if the item is selected in the response data list.

The message "The Reference

Waveform Learning Data File may

have been damaged." is displayed in

the diagnosis operation setting.

Is reference waveform learning data

that is used in the diagnosis

operation setting correct?

• Set the reference waveform learning data file again.

• Re-create the reference waveform learning data file, and set the file again.

The number of saved diagnosis result

files exceeds the set upper limit.

Can a diagnosis result file be

deleted?

Check if the diagnosis result file is being accessed.

Display language is different from the

one that has been set.

Does a user have the access right to

the folder in which Real-time Data

Analyzer is stored?

• Perform the following procedure.

Right-click the folder in which Real-time Data Analyzer is stored*5.

Select [Properties], and click the [Security] tab.

Click the [Edit] button.

Click the name of a logged in user.

Select the checkbox of "Allow" in "Permissions."

Failed to export settings. Does a user have the access right to

the folder in which Real-time Data

Analyzer is stored?


Right-click the folder in which Real-time Data Analyzer is stored*5.

Select [Properties], and click the [Security] tab.

Click the [Edit] button.

Click the name of a logged in user.

Select the checkbox of "Allow" in "Permissions."

The message "Failed to import the

settings." is displayed when importing

settings.

Is the version of Real-time Data

Analyzer the latest?

Upgrade Real-time Data Analyzer to the latest version.



17

C:\Mitsubishi Electric Edgecross Software\RDA


MEMO

107

PA

RT

3

PART 3 Real-time Statistic Diagnosis Tool

This part explains Real-time Statistic Diagnosis Tool.

Page 108 OVERVIEW

Page 114 PROCEDURE TO USE Real-time Statistic Diagnosis Tool

Page 125 Reader

Page 164 Analyzer

Page 416 Monitor


10818 OVERVIEW

18 OVERVIEW

Real-time Statistic Diagnosis Tool is an edge application that analyzes and diagnoses data in a CSV file output in Real-time

Flow Manager by using a statistical method or multivariate analysis.

The following operations are mainly performed in Real-time Statistic Diagnosis Tool.

• Reading a CSV file and creating the diagnosis roles by analyzing data in the CSV file

• Diagnosing data based on the set rules, and outputting a response data file and reporting an error by displaying a pop-up

message when an error is detected.

Precautions

• Do not use any functions and/or menu items that are not mentioned in this manual. If using, please take full responsibility

for the operations. (The operations are not guaranteed.)

• When using Real-time Statistic Diagnosis Tool, a mouse and keyboard are required.

• When a numerical value entered in the text box exceeds the maximum value or falls below the minimum value which can

be entered, the entered value will be replaced with the maximum or minimum value at the time when the focus moves to

another location.

• When an entered value is not fully displayed in the text box, it can be checked by placing the mouse cursor on it. If 256 or

more characters (single-byte) are entered, '...' is displayed for the exceeded characters.

18 OVERVIEW18.1 Configuration Software 109

18

18.1 Configuration SoftwareReal-time Statistic Diagnosis Tool consists of the following software.

The following describes the operation flow of each unit of software when cooperating with Real-time Flow Manager.

Real-time Flow Manager outputs data collected in a production site to a specified folder as a CSV file.

Reader reads the CSV file and registers it into the internal database of Real-time Statistic Diagnosis Tool.*1

Analyzer references the data registered into the internal database.

A user analyzes the data offline and set collection data to be diagnosed and diagnosis rules by using Analyzer.

Analyzer starts a diagnosis according to the set diagnosis rules.

When an error is detected, Analyzer registers it into the internal database.

Monitor monitors whether Analyzer registers an error into the internal database, then outputs a response data file to a

specified folder when the registration of an error is detected.

Real-time Flow Manager checks the output of a response data file and provides feedback when the output is detected.

*1 Data registered into the internal database can regularly be deleted. For details, refer to the following section.Page 143 Deleting read data

Software Description Reference

Reader To read a CSV file output to a specified folder and register it to the internal database. Page 125 Reader

Analyzer To analyze and diagnose data registered to the internal database by Reader. Page 164 Analyzer

Monitor To diagnose whether an error is detected by Analyzer.

After a diagnostic, the detection result is displayed on a screen and the response data file is

output.

Page 416 Monitor


(2) Real-time Statistic Diagnosis Tool

(3) Internal database

Reader

(2)

Monitor Analyzer

�

��

�

�

�

(3)(1)

11018 OVERVIEW18.1 Configuration Software

Screen configurationThe following shows the screen configurations of each unit of software.

Reader

For screen items, refer to the following chapter.

Page 125 Reader


*1 Displayed when 'Reader equipment data folder' is defined, and enabled to select after a unit definition is added.*2 Displayed when a unit definition is selected.*3 It is displayed in 'setting edit mode'.

Page 142 Editing Settings*4 It is displayed in 'CSV reading mode'.

Page 162 Switching the Operation Mode to CSV File Reading Mode


File Select unit definition*1 To select a unit definition.

Variable list export*2 To export the variable list. Page 152 Exporting the variable list

Variable list export*2 To import a CSV file exported in the [Variable list export] menu. Page 152 Importing the variable list

Switch to CSV reading mode*3 To switch the state in which a CSV file is read automatically. Page 162 Switching the Operation

Mode to CSV File Reading Mode

Switch to setting edit mode*4 To switch the state in which a CSV file is read automatically to

the state in which a unit definition can be edited.

Page 142 Editing Settings

Edit Restore Do not use this.

Redoing Do not use this.

Operation Auto delete

data setting

Do not delete it. To disable the automatic deletion of data. Page 143 Deleting read data

2 days to 360

days

To set the number of day for how many days of data to retain

when enabling the automatic deletion.

Data deletion

(manual)

Delete all data To delete all data.

Data interpolation To interpolate insufficient data or delete extra data when the

data interval of a CSV file is larger or smaller than the sampling

interval which was set in a unit definition.

Page 144 Interpolating data

Moving and deleting units To change the unit order or delete units in a batch. Page 154 Moving/Batch-deleting units


18

Analyzer


Page 164 Analyzer

11218 OVERVIEW18.1 Configuration Software


*1 The English manual appears when the display language of Real-time Data Analyzer is set to Chinese.


Disp Read data To display data saved in a CSV file in a waveform. Page 165 Displaying Read Data

Read data statistics To display the statistic of data saved in a CSV file. Page 180 Displaying Read Data Statistics

Read data frequency To display the frequency of data saved in a CSV file. Page 183 Displaying Read Data

Frequency

Simple analysis - Simple

diagnosis

SPC To diagnose data saved in a CSV file according to the

SPC rules.

Page 189 SPC

MVA To diagnose the statistic of data saved in a CSV file in

multivariate analysis (MVA).

Page 207 Multivariate Analysis (Single

Central Value Format)

Guard band To create a guard band of data saved in a CSV file

and perform a diagnostic.

Page 215 Guard Band Diagnosis

Advanced analysis Create correlation matrix To perform a correlation analysis by creating a

correlation matrix.

Page 222 Creating Correlation Matrix

Multiple regression

analysis (LMR)

To calculate the correlation among multiple variables

and predict the values of variables by multiple

regression equation.

Page 243 Multiple Regression Analysis

(LMR)

Maharanobis - Taguchi

method (MT)

To monitor the variations based on the correlation of

variations between multiple variables.

Page 250 Maharanobis - Taguchi method

(MT)

Analysis logic - diagnostic

logic operation

Logic editing To edit an analysis logic and a diagnostic logic

operation.

Page 255 Editing Logics

Logic variable setting To edit a variable used in an analysis logic and a

diagnostic logic.

Page 356 Setting logic variable

Data display after logic

execution

Data display after

execution (single)

To display the result of a logic performed. Page 365 Displaying Data after Running

Diagnostics result display Simple diagnosis result

display

To display a result of data which is currently being

diagnosed by any method of 'SPC', 'MVA', or 'Guard

band'.

Page 374 Displaying a Result of Simple

diagnosis

Diagnostic logic result

display

To display a log written by a logic. Page 382 Displaying a Diagnostic Logic

Result

Management Dataset To manage data used in Real-time Statistic Diagnosis

Tool.

Page 384 Management

Option Setting To set the following settings regarding to Real-time


Setting for data management

Setting for waveform display

Setting for logics

Page 390 Data Management Setting

Page 391 Waveform Display Setting

Page 393 Logic Setting

Help Manual To start e-Manual Viewer and display the manual.*1

Version Information To display the version information of Real-time Data

Analyzer.


18

Monitor


Page 416 Monitor

11419 PROCEDURE TO USE Real-time Statistic Diagnosis Tool19.1 Data Analysis Setting

19 PROCEDURE TO USE Real-time Statistic Diagnosis Tool

This chapter shows the procedure to perform a diagnosis by using Real-time Statistic Diagnosis Tool and Real-time Flow

Manager.

Real-time Statistic Diagnosis Tool can cooperate with Real-time Flow Manager that is installed in another industrial PC.

Operating procedure

1. Output a CSV file to be analyzed in Real-time Flow Manager, and display and analyze data saved in the output CSV file

in Real-time Statistic Diagnosis Tool to consider and create diagnosis rules.

Page 114 Data Analysis Setting

2. Output a CSV file to be diagnosed in Real-time Flow Manager, and diagnose data saved in the output CSV file in Real-

time Statistic Diagnosis Tool according to the created diagnosis rules.


To migrate settings of Real-time Statistic Diagnosis Tool to the tool in another industrial PC, refer to the

following:

Page 121 Migrating settings to Real-time Statistic Diagnosis Tool in another industrial PC

19.1 Data Analysis SettingThis section shows the procedure to create a diagnosis rule in Real-time Statistic Diagnosis Tool by using a CSV file output in

a data logging flow in Real-time Flow Manager.

Operating procedure

1. Define a folder used in Real-time Statistic Diagnosis Tool by using Reader.

Page 129 Folder Definition

2. Create a data logging flow setting by using Real-time Flow Designer.


3. Perform a data logging flow in the created data logging flow setting.

For performing a data logging flow, refer to the following:


4. By using Reader, define the format of a CSV file to be read and output a response data definition file.

Page 131 Adding a Unit Definition

5. By using Reader, set the settings to read a CSV file automatically in Real-time Statistic Diagnosis Tool.


6. Display and analyze the read CSV file and create the diagnosis rules by using Analyzer.

Page 164 Analyzer

19 PROCEDURE TO USE Real-time Statistic Diagnosis Tool19.1 Data Analysis Setting 115

19

Creating the data logging flow settingCreate the data logging flow setting by using Real-time Flow Designer.



Operating procedure


2. Click "Data Logging Flow Setting" in the edit item tree.


The "Data Logging Flow Setting" screen appears.

4. Set the setting for data collection.

5. Set the data modification setting as necessary.

When setting the data modification setting, steps 6 to 8 are unnecessary.

6. Select "Data Extraction" for "Function Type" for data modification, and click the "Detailed Setting" column.

7. Set any of data set for "Input Data" to "Monitor Target" and "Comparison Target", and set "=" for "Comparison Condition".


11619 PROCEDURE TO USE Real-time Statistic Diagnosis Tool19.1 Data Analysis Setting

9. Set the data storing setting for the last data modification process.


*1 When specifying "Hexadecimal Format", it is regarded as a character string in Real-time Statistic Diagnosis Tool.*2 If date and time data cannot be recognized, it is converted into '1904/01/01 00:00:00' in Real-time Statistic Diagnosis Tool.

Item Consideration

Executable form Select "File".

File Format Select "CSV file (Shift_JIS)".

Output Format Specify it according to the data type of data to be output as follows:

• The data type is UINT, UDINT, or ULINT: Specify "Integer Format".*1

• The data type is REAL or LREAL: Specify "Decimal Format" (set the

number of digits after the decimal point up to 7 digits) or "Index Format".

Date and time format Set any of the following:*2

• YYYY/MM/DD hh:mm:ss.000 ('000' is optional.)

• YYYYMMDDhhmmss000 ('000' is optional.)

Save Destination Folder Set 'UNIT' folder in the CSV input folder defined in the following section.


Prefix of file name Set a string from 1 to 12 characters or that of 14 or more characters.

19 PROCEDURE TO USE Real-time Statistic Diagnosis Tool19.2 Data Diagnosis Setting 117

19

19.2 Data Diagnosis SettingThis section shows the procedure to perform a diagnosis according to the diagnosis rule by using a CSV file to be diagnosed

that is output in a data diagnosis flow in Real-time Flow Manager.

Real-time Statistic Diagnosis Tool outputs a response data file when an error is detected in a diagnosis.

Real-time Flow Manager detects the output of a response data file and feeds back which is set in Real-time Flow Designer.

Precautions

To cooperate with Real-time Flow Manager that is installed in another industrial PC, set a storage destination folder for a

response data file in the industrial PC ((installation destination of Edgecross Basic Software)\Edgecross\Edgecross Basic

Software\RDDIF_Output) as a shared folder which can be read and written.

Operating procedure

1. Create a data diagnosis flow setting by using Real-time Flow Designer.


2. By using Monitor, set the settings to output a response data file when an error is detected in a diagnosis.

Page 416 Monitor

To cooperate with Real-time Flow Manager that is installed in another industrial PC, specify a response data

file output folder by using Reader.

Page 120 Cooperation with Real-time Flow Manager installed in another industrial PC

3. By using Analyzer, start a diagnosis.

Page 164 Analyzer

4. Perform a data diagnosis flow in the created data diagnosis flow setting.

For performing a data diagnosis flow, refer to the following:


After starting an industrial PC, Real-time Statistic Diagnosis Tool cannot automatically turn into a state in

which a diagnosis can be performed. Starting Real-time Statistic Diagnosis Tool and performing 'Switch to

CSV file reading mode' in Reader are required. (Page 162 Switching the Operation Mode to CSV File

Reading Mode)

11819 PROCEDURE TO USE Real-time Statistic Diagnosis Tool19.2 Data Diagnosis Setting

Creating the data diagnosis flow settingCreate a data diagnosis flow setting by using Real-time Flow Designer.



Operating procedure


2. Click "Data Diagnosis Flow Setting" in the edit item tree of Real-time Flow Designer.

3. Click the [Move from Data Logging Flow] button.

4. Select a flow set in "Data Logging Flow Setting" and click the [] button.


The flow set in "Data Logging Flow Setting" moves to "Data Diagnosis Flow Setting".

19 PROCEDURE TO USE Real-time Statistic Diagnosis Tool19.2 Data Diagnosis Setting 119

19


The "Data Diagnosis Flow Setting" screen appears.

7. Click the "Data Storing" column of the last data modification process.

8. Select "Do not execute" for "Executable form" in the [Executable form] tab.


10.Select "Edge Application Diagnosis (file)" in the "Function Type" column of a data diagnosis, then click the cell of the

"Detailed Setting" column.

11. Set the settings for the [Output Data] tab, [Output Option] tab, and [Save File] tab of the edge application diagnosis (file)

setting same as the data storing setting set in the following section.


12.Click the [Get Response Result Definition] button in the [Response Data] tab of the edge application diagnosis (file)

setting to select a response data definition file that is output in the following section.


13.Click the [OK] button.

14.Set the settings for feedback.

15.Click the [OK] button.

12019 PROCEDURE TO USE Real-time Statistic Diagnosis Tool19.2 Data Diagnosis Setting

Cooperation with Real-time Flow Manager installed in another industrial PCTo cooperate with Real-time Flow Manager that is installed in another industrial PC, specify a response data file output folder

by using Reader.

Operating procedure

1. Select the checkbox of "Link to the basic software of other IPC."

2. Specify the storage destination folder ((installation destination of Edgecross Basic Software)\Edgecross\Edgecross Basic

Software\RDDIF_Output)*1 for a response data file in "Response data file output folder."

*1 Specify a folder path which can be recognized in Windows.

Precautions

• In the following case, the items of "Link to the basic software of other IPC" are displayed behind the variable list as shown

below: Real-time Data Analyzer whose version is 1.05F or later is installed in an industrial PC in which the Real-time Data

Analyzer whose version is 1.04E or earlier has been installed before.

If so, adjust the width of the variable list.

• When the capacity of an output destination drive for a response data file is insufficient, the output stops. It will start again

when the insufficiency is resolved. (Response data output while the capacity is insufficient is discarded.)

• When setting a folder path in "Response data file output folder," set the path so that the path length of a response data file

to be output does not exceed 259 characters. For the name of a response data file, refer to the following:

Page 16 Real-time Statistic Diagnosis Tool

• When the folder set in "Response data file output folder" is changed, restart Analyzer and Monitor.

19 PROCEDURE TO USE Real-time Statistic Diagnosis Tool19.3 Migrating settings to Real-time Statistic Diagnosis Tool in another industrial PC 121

19

19.3 Migrating settings to Real-time Statistic Diagnosis Tool in another industrial PC

This section shows the procedure for migrating settings of Real-time Statistic Diagnosis Tool to the tool in another industrial

PC.

Exporting and importing a setting file (extension: rsd) is required by using the project file setting utility. (Page 122 Project

File Setting Utility)

Operating procedure

1. Perform the following procedure in a migration source industrial PC.

Page 121 Procedure to be performed in a migration source industrial PC

2. Perform the following procedure in a migration destination industrial PC.

Page 121 Procedure to be performed in a migration destination industrial PC

Procedure to be performed in a migration source industrial PC1. Check that a diagnosis can be performed normally in Real-time Statistic Diagnosis Tool.

2. Start the project file setting utility.

Page 122 Startup method

3. Export a setting file.

Page 122 Export procedure

Procedure to be performed in a migration destination industrial PC1. Install Real-time Data Analyzer.

This step is not necessary if migrating settings to an industrial PC in which it is already installed.

2. Use the utility for switching language, and set the same language as the display language of the migration source Real-

time Data Analyzer.


3. Start Real-time Statistic Diagnosis Tool.


4. Set each folder in Reader according to the environment of the migration destination industrial PC.


5. Start the project file setting utility.

Page 122 Startup method

6. Import the setting file exported in the migration source industrial PC.

Page 124 Import procedure

If the settings of Real-time Statistic Diagnosis Tool have been configured, its current settings will be

overwritten by importing a setting file. It is recommended to export the current settings for a backup before

importing the file.

7. Select the exported unit for "Switch Pattern" in the dataset settings of Analyzer.

Page 384 Management

12219 PROCEDURE TO USE Real-time Statistic Diagnosis Tool19.4 Project File Setting Utility

19.4 Project File Setting UtilityThis section shows the procedure to export and import a setting file by using the project file setting utility.

Startup method

Operating procedure


2. Double-click 'ProjectFile Utility.'

The display language of the project file setting utility follows the language set in the utility for switching

language.


Considerations • The display language in Real-time Statistic Diagnosis Tool needs to be the same when exporting and importing a file.

• If file contents had been set in a language different from the one at the time of export, those contents cannot be exported.

• Only the settings stored in 'Reader equipment data folder' or 'RSD dataset folder' can be exported; however, the following

settings are not exported.

Settings for creating a correlation matrix (Page 222 Creating Correlation Matrix)

Data management settings (Page 390 Data Management Setting)

Monitor settings (Page 419 Monitor Setting)

• A setting file cannot be exported when the save destination folder for the file saved by clicking any of the following buttons

is not 'Reader equipment data folder' or 'RSD dataset folder.' In this case, move the file manually to the migration

destination as necessary.

Export procedure

Operating procedure

1. Select [Export].

2. To export data that has already been read*1, select the checkbox of "Export the data that has been read."

*1 If so, unit data can be recovered in the Real-time Statistic Diagnosis Tool to which the setting file is migrated. (Page 159 Recovering unit data)

Screen Button Reference

Read data display • [Save] button

• [CSV for matrix] button

Page 165 Displaying Read Data

Read data statistics display • [Save] button


Page 180 Displaying Read Data Statistics

Read data frequency display • [Save] button


Page 183 Displaying Read Data Frequency

SPC • [Save] button


Page 189 SPC

Multivariate analysis (single central value

format)

• [Save] button


• [Save criterion.] button

Page 207 Multivariate Analysis (Single Central Value

Format)

Guard band diagnosis • [Save] button


• [Save GB] button


19 PROCEDURE TO USE Real-time Statistic Diagnosis Tool19.4 Project File Setting Utility 123

19

3. Click the [Next] button.


5. Specify the save destination folder path of the file to be exported to and the file name*1, then click the [Save] button.

*1 The number of characters that can be specified is as follows:Folder path: up to 247 charactersFile path: up to 259 characters

• File export can be cancelled by clicking the [Cancel] button in the screen displayed while exporting a file.

• Standard space required for export is as follows. (Required space depends on a setting file.)

(Size of 'Reader equipment data folder' + size of 'RSD dataset folder') 2

12419 PROCEDURE TO USE Real-time Statistic Diagnosis Tool19.4 Project File Setting Utility

Import procedure

Operating procedure

1. Select [Import].

2. To export the current Real-time Statistic Diagnosis Tool settings before import, select the checkbox of "Export the current

settings before importing."

To export data that has already been read as well*1, select the checkbox of "Export read data."

*1 If so, unit data can be recovered in the Real-time Statistic Diagnosis Tool to which the setting file is migrated. (Page 159 Recovering unit data)



5. Select the setting file to be imported*1, then click the [Open] button.

*1 The number of characters that can be specified is as follows:Folder path: up to 247 charactersFile path: up to 259 characters

6. When selecting the checkbox of "Export the current settings before importing.," perform steps 4 and 5.

Page 122 Export procedure

• When performing step 6, a setting file is imported after exported.

• File export or import can be cancelled by clicking the [Cancel] button in the screen displayed while

exporting/importing a file; however, note that it may take time to roll back the settings when cancelling the

import.

• Standard space required for export is as follows. (Required space depends on a setting file.)

(Size of 'Reader equipment data folder' + size of 'RSD dataset folder') 2

• Standard space required for import is as follows. (Required space depends on a setting file.)

(Size of an import target setting file 9) + (size of 'Reader equipment data folder' + size of 'RSD dataset

folder') 2

20 Reader 125

20

20 Reader

This software reads a CSV file that was output to a specified folder.

The following shows the operating procedure of Reader.

Operating procedure

1. Define a folder used in Reader.


2. Add a definition of a CSV file to be read.


3. Edit a setting.

Page 142 Editing Settings

4. Switch the status to the one for reading a CSV file automatically.


12620 Reader20.1 Screen Configuration

20.1 Screen ConfigurationThe following shows the screen configuration of Reader.

Displayed items

*1 Do not specify a folder with the encryption attribute enabled.*2 It can be specified only when the checkbox of "Link to the basic software of other IPC" is selected.

Item Description

(1) Log list Time A time when a log is output is displayed.

LogComment An output log is displayed.

For details on a log, refer to the following:

Page 127 Log list displayed in LogComment

Reader equipment data folder*1 Specify the path of a folder storing the settings, data, and others for Reader (up to 256

characters).

CSV input folder*1 Specify the path of a folder storing a CSV file to read in Reader (up to 256 characters).

RSD dataset folder*1 Specify the path of a folder storing the settings and data for Analyzer, data to diagnose in Monitor,

and others (up to 256 characters).

Link to the basic software of other IPC Select this checkbox to cooperate with Real-time Flow Manager that is installed in another

industrial PC.

Response data file output folder*1,*2 Specify the folder to store a response data file ((installation destination of Edgecross Basic

Software)\Edgecross\Edgecross Basic Software\RDDIF_Output).

(2) Variable list Name A variable name is displayed.

It can be changed when performing 'Switch to setting edit mode'.

Page 160 Changing a variable name

Type (numerical value |

char)

The data type of a variable is displayed.

• String: Text variable. Text data which changes in each section and can be used for filtering

sections to display.

• Numeric: Trace variable. Numerical value data which changes continuously.

• varn: Step variable. Data which does not change continuously and can be used to extract

sections.

The data type can be changed when performing 'Switch to setting edit mode'.

Page 160 Changing a data type

Unit Do not use this.

Expression An operational expression can be entered when performing 'Switch to setting edit mode'.

The operation result can be used as data in Analyzer.

Page 161 Entering an operational expression

Limit pair diagnostics Do not use this.

(1)

(2)

20 Reader20.1 Screen Configuration 127

20

Log list displayed in LogCommentThe following table shows the contents displayed in the "LogComment" column.

*1 If reading CSV file data fails, the data is discarded.

Item Description

File was read. Displayed when CSV data was read successfully.

File was read. (There is an interpolation) Displayed when CSV data was read successfully and interpolated.

Failed to read file.*1 Displayed when CSV data stored in the CSV input folder was deleted while the data was being read and in

other cases.

Because the time stamp that overlapped with

existing data existed, data was not updated.

Displayed when CSV data with the time stamp duplicate with or earlier than that of already-read CSV data

was added.

CSV Data is saved in the following folder under 'Reader equipment data folder'.

• Reader_UNIT\YYMMDD\Original\Invalid

('YYMMDD' indicates the current date.)

It became invalid because it was an undefined

file.

Displayed when a CSV file which was not specified in the unit definition was read.

When this log is displayed, the invalid file is saved in the following folder under 'Reader equipment data

folder'.



While editing Displayed when the change of the unit setting was started in the setting edit mode.

Content of the edit was saved. Displayed when the [Save change] button was clicked.

Edit was canceled. Displayed when the [Cancel] button was clicked.

Delete unit data - (YYmmss) Start Displayed when a deletion of the latest unit data specified in the unit data deletion operation was started.

Delete unit data - (YYmmss) End Displayed when a deletion of the latest unit data specified in the unit data deletion operation was ended.

Delete unit data - Start Displayed when a deletion of all unit data specified in the unit data deletion operation was started.

Delete unit data - End Displayed when a deletion of all unit data specified in the unit data deletion operation was ended.

Unit data recovery - Start Displayed when unit data recovery was specified.

Unit data recovery - Original CSV files have been

moved into loading folder.

Displayed when unit data recovery was specified and the recovery data was stored in 'CSV input folder'.

Delete data - Delete all data - Start Displayed when a manual deletion of all data was started.

Delete data - Delete all data - Finish Displayed when a manual deletion of all data was ended.

Delete data - YYmmdd Auto Displayed when data was deleted automatically according to the automatic data deletion setting.

It became invalid because sampling interval of

CSV file is incorrect.

Displayed when the difference between the sampling interval which is set in the unit definition and that of

the CSV file to be read is large.

When this log is displayed, the invalid file is saved in the following folder under 'Reader equipment data

folder'.



12820 Reader20.2 Right-click Menu Items

20.2 Right-click Menu ItemsThis section shows the right-click menus.

Right-clicking on the variable list

Window

Displayed items

Item Description

Select unit definition To select a unit definition.

Add unit definition To add a unit definition.

Change unit definition To change a unit definition.

Delete unit definition To delete a unit definition that is added last.

Add variable To add the variables of a unit added in 'Adding a Unit Definition.'

Variable list export To export the variable list.

Variable list import To import a CSV file exported in 'Exporting the variable list.'

Edit item To edit items in the variable list such as a name, type (numerical value | character), and

operation expression.

String To change the current type (numerical value | character).

Numeric

varn

Delete unit data To delete read data for each unit.

Unit data recovery To re-read all backup data of a CSV file.

Extended section definition Add section definition To define an extended section.

20 Reader20.3 Folder Definition 129

20

20.3 Folder DefinitionThe following table shows the folder definitions used in Reader.

*1 To cooperate with Real-time Flow Manager that is installed in another industrial PC, refer to the following:Page 130 Cooperation with Real-time Flow Manager installed in another industrial PC

Operating procedure

1. Specify folder paths for "Reader equipment data folder", "CSV input folder", and "RSD dataset folder" (up to 256

characters).

The paths of the folders which are created when Real-time Data Analyzer is installed are set for each field as the default. Any

paths can be set but the default paths are preferred under normal use.

2. Select [File] [Switch to CSV file reading mode].

Subfolders necessary for operation of Real-time Statistic Diagnosis Tool are created in a specified folder.

By storing a CSV file which is output by Real-time Flow Manager into the 'UNIT' folder under 'CSV input

folder', Reader can read the file automatically.

Precautions

• For "Reader equipment data folder" and "RSD dataset folder," only files in local storage (built-in storage or eternal media)

can be specified. Folders on a network or shared folder cannot be specified.

• When specifying an external media for "Reader equipment data folder" or "RSD dataset folder," format the media in NTFS

format.

For checking if the media is formatted in NTFS format, follow the procedure below:

Right-click a target drive and select [Property].

Check that "NTFS" is set for "filesys."

• Do not change the drive letter (DE, for example) after starting a diagnosis. When the letter needs to be changed, set the

diagnosis rules again after changing the letter.

Item Description

Reader equipment data folder*1 Folder storing the settings, data, and others for Reader

CSV input folder*1 Folder storing a CSV file to be read in Reader

RSD dataset folder Folder storing the settings and data for Analyzer, data to be diagnosed in Monitor, and others

13020 Reader20.3 Folder Definition

• For "Reader equipment data folder," "CSV input folder," "RSD dataset folder," and "Response data file output folder," do not

specify a folder and subfolders that are already set in another item or a folder with the encryption attribute enabled.

• When the path set in "Reader equipment data folder," "CSV input folder," or "RSD dataset folder" is changed, restart

Analyzer and Monitor.

• When installing over Real-time Data Analyzer or updating it, the folders specified by a user before the overwrite installation

or update will be set for "Reader equipment data folder," "CSV input folder," and "RSD dataset folder."

• When setting a folder path in "CSV input folder," set the path so that the path length of a file to be stored in the 'UNIT' folder

under 'CSV input folder' does not exceed 259 characters.

• When setting a folder path in "Reader equipment data folder" and "RSD dataset folder," set the path so that the path length

of a setting file to be created in the set folder does not exceed 259 characters. The maximum path length of a setting file to

be created in the set folder is as follows:

Cooperation with Real-time Flow Manager installed in another industrial PCTo cooperate with Real-time Flow Manager that is installed in another industrial PC, set "CSV input folder" and "Reader

equipment data folder" as follows.

• CSV input folder: Specify a shared folder*1, which can be read and written, on an industrial PC in which Real-time Flow

Manager is installed.

*1 Specify a folder path which can be recognized in Windows.

• Reader equipment data folder: Specify a folder that can be referred from an industrial PC in which Real-time Flow Manager

is installed.

Precautions

• The processing time may take longer than when cooperating with Real-time Flow Manager installed in a same industrial PC

since detection of a read target CSV file requires time.

• If a read target CSV file became undetectable due to the network disconnection, the read operation will start again when

the network recovers. However, it may take time depending on the operation environment.

Item Maximum path length of a setting file

Reader equipment data folder (Folder set in the item shown left)\ + 100 characters (If the unit name and diagnosis target variable name are 32

characters long each)

RSD dataset folder (Folder set in the item shown left)\ + 187 characters (If the unit name and diagnosis target variable name are 32

characters long each)

20 Reader20.4 Adding a Unit Definition 131

20

20.4 Adding a Unit DefinitionThis section shows the operating procedure for specifying a sample CSV file, defining the CSV file to be read in Reader (unit

definition), and outputting a response data definition file.

A unit needs to be defined for each process flow defined in Real-time Flow Manager.

Reader extracts a section (unit section) from CSV files and reads the data within the section.

Operating procedure

1. Select [File] [Switch to setting edit mode].

2. Click the [Yes.] button.

3. Right-click on the variable list, and select [Add unit definition].

4. Set each item for a sample CSV file and click the [Next] button.

For the considerations for the setting, refer to the following:

Page 136 Considerations for setting a sample CSV file

*1 A file in a folder with the encryption attribute enabled cannot be specified.*2 When the name format of a CSV file to be read is 'LOG001_NNNNNNNN.csv', specify it as 'LOG001*.csv'.*3 '*' (asterisk) is used as a wild card indicating any 0 or more characters.*4 It is automatically set according to the file name set for "Reading target CSV file name". At the time, unavailable characters for a unit

name are removed. Modify the unit name as necessary.For unavailable characters for a unit name, refer to the following:Page 436 Real-time Statistic Diagnosis Tool

Item Description

Sample CSV file*1 Specify the path of a CSV file in the same format as a CSV file to be read in Reader (up to 256

characters).

Reading target CSV file name*2,*3 Specify the name of a CSV file to be read in Reader by using the wild card (*) (up to 256

characters).

Unit name*4 Enter the same character string as that set for "File Name Prefix" in the data storing setting (file

saving) in Real-time Flow Designer (up to 32 characters).


13220 Reader20.4 Adding a Unit Definition

5. Change the type of data for each row as necessary, and click the [Next] button.*1

*1 For a CSV file output in Real-time Flow Manager, changing the type is not necessary since it is set automatically.

For the considerations on changing the data type, refer to the following:

Page 136 Considerations for setting a type

• By clicking a row header (leftmost column), the type of data on the clicked row can be changed. (It can be

changed only to "Row item title," "Data," or "Unused.")

• For the first row, the type of data is set to "Row item title" by default. If the type for the first row in the CSV

file to be read is not "Row item title," change it.

• For the "Row item title" row, specify the row where the title of each column (variable name) is described.

• Only one row can be specified as "Row item title."

6. Change the data type of data for each column as necessary, and click the [Next] button.

For the considerations for changing column items, refer to the following:

Page 136 Considerations for the settings of column items

• By clicking a column header (top row), the attribute of data on the clicked column can be changed.

• Time stamps are displayed in the first column by default. Change the attribute if the first column of a CSV

file to read is not for time stamps.

• Only one time stamp column is necessary.

• Up to 5000 column items (variables) can be defined for a CSV file.

7. Set each item for the start row of data reading and click the [Next] button.

Item Description

Starting condition type Set the data extraction setting (start row).

For details on the setting, refer to the following:

Page 137 Section start conditions

Text extraction base time Select the time to extract a text variable value either "After the section starts" or "Before end of

section".

Set this when setting a value other than that of the first record as a text variable value.

Page 139 Extraction setting of a text variable value for one section


20

The section start is registered in "Dataset" as an event (time stamp of the occurrence time) with the following

name:

• (Unit name)_Start

Events can be used in particular primitives defined in 'Logic editing'.


For details on datasets, refer to the following:

Page 384 Management

8. Set each item for the end row of data reading and click the [Next] button.

Page 136 Considerations for setting "End condition type"

The section end is registered in "Dataset" as an event (time stamp of the occurrence time) with the following

name:

• (Unit name)_End

Events can be used in particular primitives defined in 'Logic editing'.


For details on datasets, refer to the following:

Page 384 Management

9. Set each item for sampling interval and click the [Next] button.

Page 136 Considerations for setting sampling interval

Offset Set an offset from the time set for "Text extraction base time".

Set a positive value when "After the section starts" is selected for "Text extraction base time" and a

negative value when "Before end of section" is selected.


• When "sec" is selected for "Unit", set this in the range from -86399.9999 to 86399.9999.

• When "%" is selected for "Unit", set this in the range from 0 to 100.

Unit Select either "sec" (second) or "%" for the offset unit. ("%" is for the section length.)

Item Description

End condition type Set the data extraction setting (end row).


Page 140 Section end conditions

Equipment common shortest section length (sec) Set the shortest section length in seconds (0.0000 to 2147483647.0000).

If the section end condition is satisfied before the specified length is reached, the section is

discarded.

Equipment common longest section length (sec) Set the longest section length in seconds (0.0000 to 2147483647.0000).

If the section end condition is not satisfied when the specified length is reached, the section is

discarded.

Item Description


Data interpolation can be performed when data sampled at different intervals are contained.

Page 144 Interpolating data

10.Click the [Save change] button.

A response data definition file ((unit name)_DATATYPE.csv) is created in the following folder when clicking the [Save change]

button.

• (Folder set in "Reader equipment data folder")\DATATYPE*1

*1 If there is no DATATYPE folder, it will be created at the same time when a response data definition file is created.


The set unit definition is saved.

Item Description

Sampling interval Set the data sampling interval of a CSV file.

Set a value and a unit (seconds, milliseconds, or microseconds).

• Setting range for 'seconds': 1 to 86400

• Setting range for 'milliseconds': 1 to 86400000

• Setting range for 'microseconds': 1 to 1000000000

Default retrieval period Set the default display period in the fields of "From" and "To" in each operation screen of the

following functions.




Page 189 SPC

Page 207 Multivariate Analysis (Single Central Value Format)


When changing a value in "From", the date and time in "To" is automatically set (0 to 4294967295)

by adding the time which is set in this option to the value in "From".


20

12.Repeat the above procedure to add another unit definition if any.

• To discard the set unit definition, click the [Cancel] button.

• Up to 99 unit definitions can be added.

The following restrictions apply when reading a CSV file the data sampling interval of which is set in

microseconds or nanoseconds.

• Data in nanoseconds cannot be read.

• In the [Analysis logic - diagnostic logic operation] or [Data display after logic execution] menus of Analyzer,

data in less than milliseconds cannot be used.

• In the [Disp] or [Simple analysis - Simple diagnosis] menus of Analyzer, data in microseconds or more can

be used, but it is rounded off to milliseconds on the display.


Precautions

■ Considerations for setting a sample CSV file • A read-only file cannot be specified as a sample CSV file.

• The CSV file format cannot be set if the file is exclusively controlled such as being opened in a different tool and cannot be

written.

• The same name as an existing unit already registered cannot be defined.

• Do not rename a unit after it has been defined and data has been acquired. If it is renamed, the unit data acquired before

renaming can no longer be referred by Analyzer.

■ Considerations for setting a type • A file including variables whose names are duplicated cannot be specified as a sample CSV file.

• The following characters cannot be used in a variable name:

\ (0x5C), / (0x2F), : (0x3A), * (0x2A), ? (0x3F), " (0x22), < (0x3C), > (0x3E), | (0x7C)

￥ (0x818f), ／ (0x815e), ： (0x8146), ＊ (0x8196), ？ (0x8148), ” (0x8168), ＜ (0x8183), ＞ (0x8184), ｜ (0x8162)

■ Considerations for the settings of column items • For reading a column item which was set to "Unused", add the item to the last column of the CSV file and define that file

again.

• A new "Unused" column item cannot be set in version 1.08J or later since "Unused" is no longer displayed in the right-click

menu for the column header. The column items that were set to "Unused" in version 1.07F or earlier are displayed as

"Unused"; however, the setting cannot be changed since the right-click menu is not displayed.

■ Considerations for setting "End condition type" • Set the section length to at least one second or longer. Otherwise, the data in the created section cannot be used correctly.

■ Considerations for setting sampling interval • In "Sampling interval", the average sampling interval of a sample CSV file is displayed as the default. Modify it according to

the actual sampling interval as necessary.

• Do not set the sampling interval to a value which is greatly differs from the actual sampling interval. Otherwise, an error may

occur and Real-time Statistic Diagnosis Tool may not work properly.

• Set the sampling interval of a CSV file as constant as possible.

• When the sampling intervals of a CSV file vary, set the average of the actual sampling interval in "Sampling interval".

• Set the section length to shorter than one day (86400 seconds). If a single section spans one day or longer, data in this

section cannot be used in Analyzer.


20

Section start conditionsThe following three conditions are available for the start of a unit section and an extended section.

Precautions

When the start condition is satisfied multiple times before the end condition is satisfied, the last satisfied start condition is

used.

■ When setting the section start time to the time of the time stamp at the first row

Operating procedure

1. Select "Time of the first line" for "Starting condition type".

For an extended section, "Offset (sec)" can also be specified.

2. The section start position is set to the first row of the CSV file.

■ When setting the section start time to the timing that a text matches, mismatches, or changes to another value

Operating procedure

1. Select "Text" for "Starting condition type".

2. Select a variable for "Variable name". An option the data type of which is "Text" can be selected.

3. Select "= (equal)", "!= (not equal)", or "Changing point" for "Operator".

4. Enter a text to compare in "Comparison value" (up to 32 characters).


5. The section start position of each option is shown below.

• "=(equal)":

Row where there is a text which matches the text entered in "Comparison value"

• "!=(not equal)":

Row where there is no text which matches the text entered in "Comparison value"

• "Changing point":

Row where the text changes to another one

• When not setting a text as a condition, it is not necessary to enter a text in "Comparison value".

• To set multiple texts with OR condition, enter the values delimited by a comma in "Comparison value".

• When "Changing point" is selected, the section(s) until the current text is changed to another is discarded.

• A wild card (* or ?) can be entered in "Comparison value".

The wild card '*' represents two or more characters and '?' represents a single character.


■ When setting the section start time to the timing that a numerical value matches, mismatches, exceeds or falls below the threshold

The section start position of each option is shown below.

Operating procedure

1. Select "Numeric" for "Starting condition type".

2. Select a variable for "Variable name". An option the data type of which is "Numeric" can be selected.

3. Select "= (equal)", "!= (not equal)", "UP(Less than TO greater than or equal)", or "DOWN(Greater than TO less than or

equal)" for "Operator".

4. Enter a value to compare in "Comparison value" (up to 32 characters).

5. To add a condition, enter it in "Additional AND condition".


6. The section start position of each option is shown below.

• "=(equal)":

Row where there is a variable value which matches the value entered in "Comparison value"


Row where there is no variable value which matches the value entered in "Comparison value"

• "UP(Less than TO greater than or equal)":

Row where a variable value exceeds the value entered in "Comparison value"

• "DOWN(Greater than TO less than or equal)":

Row where a variable value falls below the value entered in "Comparison value"

• To set multiple numerical values with OR condition, enter the values delimited by a comma in "Comparison

value".

• Up to three conditions can be added.

Precautions

For "!= (not equal)", when the set variable changes in a section, the time of the last change is set as the section start time.


20

Extraction setting of a text variable value for one sectionRegardless of how many values are included in a section, only one value can be retained as a text variable value in one

section.

By default, the value of the first record in a section is used as a text variable value. However, another value can be used by

setting each item of "Text extraction base time" and "Offset".

Precautions

When the following settings are set and if no data is recorded at the time corresponding to the content set for "Offset", a text

variable value will not be set.

• "Text extraction base time" is set to "After the section starts" and a negative value is set for "Offset".

• "Text extraction base time" is set to "Before end of section" and a positive value is set for "Offset".

(1) When "Text extraction base time" is set to "After the section starts"

(2) When "Text extraction base time" is set to "Before end of section"

(a) Offset from the start row (A positive value is set for "Offset".)

(b) Offset from the end row (A negative value is set for "Offset".)

2017/7/1 3:00:002017/7/1 3:00:022017/7/1 3:00:042017/7/1 3:00:062017/7/1 3:00:082017/7/1 3:00:10

ABCCCC

101102103104105106

(1)

(a)

(2)2017/7/1 3:00:002017/7/1 3:00:022017/7/1 3:00:042017/7/1 3:00:062017/7/1 3:00:082017/7/1 3:00:10

ABCCCC

101102103104105106(b)


Section end conditionsThe following four conditions are available for the end of a unit section and an extended section.

■ When setting the section end time to the time of the time stamp at the last row

Operating procedure

1. Select "Time of the last line" for "End condition type".

For an extended section, "Offset (sec)" can also be specified (negative value only).

2. The section end position is set to the last row of the CSV file.

■ When setting the section end time to the timing that a text matches, mismatches, or changes to another value

Operating procedure

1. Select "Text" for "End condition type".

2. Select a variable for "Variable name". An option the data type of which is "Text" can be selected.

3. Select "= (equal)", "!= (not equal)", or "Changing point" for "Operator".

4. Enter a text to compare in "Comparison value" (up to 32 characters).


5. The section end position of each option is shown below.

• "=(equal)":

Row where there is a text which matches the text entered in "Comparison value"


Row where there is no text which matches the text entered in "Comparison value"

• "Changing point":

Row where the text changes to another one

• When not setting a text as a condition, it is not necessary to enter a text in "Comparison value".

• To set multiple texts with OR condition, enter the values delimited by a comma in "Comparison value".

• A wild card (* or ?) can be entered in "Comparison value".

The wild card '*' represents two or more characters and '?' represents a single character.


20

■ When setting the section end time to the timing that a numerical value matches, mismatches, exceeds or falls below the threshold

Operating procedure

1. Select "Numeric" for "End condition type".

2. Select a variable for "Variable name". An option the data type of which is "Numeric" can be selected.

3. Select "= (equal)", "!= (not equal)", "UP(Less than TO greater than or equal)", or "DOWN(Greater than TO less than or

equal)" for "Operator".

4. Enter a value to compare in "Comparison value" (up to 32 characters).

5. To add a condition, enter it in "Additional AND condition".


6. The section end position of each option is shown below.

• "=(equal)":

Row where there is a variable value which matches the value entered in "Comparison value"


Row where there is no variable value which matches the value entered in "Comparison value"

• "UP(Less than TO greater than or equal)":

Row where a variable value exceeds the value entered in "Comparison value"

• "DOWN(Greater than TO less than or equal)":

Row where a variable value falls below the value entered in "Comparison value"

• To set multiple numerical values with OR condition, enter the values delimited by a comma in "Comparison

value".

• Up to three conditions can be added.

Precautions

When setting "= (equal)" for "Operator" and '0' for "Comparison value", the time corresponding to the last row of the CSV file

is set as the section end time.

■ When setting the section end time to the timing that the next section starts

Operating procedure

1. Select "Until the start of the next section" for "End condition type".

2. The first row of the next section is set to the end position of the current section.

14220 Reader20.5 Editing Settings

20.5 Editing SettingsSettings can be edited as necessary.

The following table shows the available operations.


Deleting read data To delete the data in a read CSV file. Page 143 Deleting read data

Interpolating data To interpolate insufficient data or delete extra data when the data

interval of a CSV file is larger or smaller than the sampling interval

which was set in a unit definition.

Interpolating data

Changing a unit definition To change a unit setting. Page 146 Changing a unit definition

Deleting a unit definition To delete the last added unit. Page 147 Deleting a unit definition

Adding variables To add the variables of a unit added in 'Adding a Unit Definition.' Page 148 Adding variables

Exporting the variable list To export the variable list for modifying it significantly. Page 152 Exporting the variable list

Importing the variable list To import a CSV file which was exported by using the [Variable list

export] menu.

Page 152 Importing the variable list

Deleting unit data To delete read data for each unit. Page 153 Deleting unit data

Moving/Batch-deleting units To change the unit order or delete units in a batch. Page 154 Moving/Batch-deleting units

Defining an extended section To register the extraction method of a different section for a CSV file for

which a section definition is set at a unit definition.

Page 155 Defining an extended section

Recovering unit data To re-read all backup data of read CSV file. Page 159 Recovering unit data

Changing a variable name To change a variable name. Page 160 Changing a variable name

Changing a data type To change the data type of a variable. Page 160 Changing a data type

Entering an operational expression To enter an operational equation and use the operation result in

Analyzer.

Page 161 Entering an operational

expression

20 Reader20.5 Editing Settings 143

20

Deleting read dataTo prevent read CSV file data from occupying too much disk space, delete the CSV file data registered into the internal

database automatically or manually.

Files in the folders set for "Reader equipment data folder" and "RSD dataset folder" can be deleted.

■ Deleting data automatically

Operating procedure

1. Select [Operation] [Auto delete data setting]. Select the number of days to store data in the internal database from [2

days] to [360 days].

To set not to delete data automatically, select [Do not delete it.].

Precautions

• Note that the number of days is not the number of calendar days.

• When a new CSV file is read after 00:00:00, all files of which the storage period has expired are deleted.

Ex.

When "2 days" is selected and the following files are stored:

File (2017/01/02)

File (2017/01/09)

File (2017/01/17)

File (2017/01/23)

When a new CSV file is read at 2017/1/30 00:00:01, the files dated 2017/01/02 and 2017/01/09 are deleted.

• When a section spans more than one day, the data is used as one with the time stamp of the last date in the section.

Ex.

When the date of data in a section is from 2017/01/11 23:50:00 to 2017/01/12 00:10:00, it is used as the data on 2017/01/12.

The start date of the section is 2017/01/11, but the data is excluded when the deletion of other data dated 2017/01/11.

■ Deleting data manually

Operating procedure

1. Select [Operation] [Data deletion (manual)] [Delete all data].

Precautions

• By using the [Delete all data] menu, data of not only the selected unit but also all units is deleted.

• End Analyzer and Monitor before using the [Delete all data] menu.


Interpolating dataBy setting the data interpolation, when the data interval of a CSV file is larger or smaller than the value which is set for

"Sampling interval" in a unit definition, insufficient data can be interpolated or extra data can be deleted.

Data interpolation (nearest neighbor interpolation*1) or deletion is performed based on the value calculated as 'Length (time)

of one section Sampling interval'.

When there are some data with the same time stamp, the latest data among the data only is retained.

*1 A method for interpolating data by using the original data value that is closest in time

Even if the data sampling interval is not constant, data will not be interpolated as far as the value calculated by

'Length (time) of one section Sampling interval' is equal to the number of units of data.

(1) When 'Length (time) of one section Sampling interval' is greater than 'Number of units of data' (Sampling interval = 2 seconds)

(2) When 'Length (time) of one section Sampling interval' is smaller than 'Number of units of data' (Sampling interval = 2 seconds)

(3) When there are some data with the same time stamp

(a) Data is interpolated in a missing part.

(b) Unnecessary data is deleted.

(c) Data other than the latest data is deleted.

2017/7/1 3:00:002017/7/1 3:00:022017/7/1 3:00:042017/7/1 3:00:062017/7/1 3:00:082017/7/1 3:00:10

122456

101102102104105106

(a)

(b)

2017/7/1 3:00:002017/7/1 3:00:022017/7/1 3:00:032017/7/1 3:00:042017/7/1 3:00:062017/7/1 3:00:08

122345

101102102103104105

2017/7/1 3:00:10 6 106

2017/7/1 3:00:002017/7/1 3:00:022017/7/1 3:00:032017/7/1 3:00:042017/7/1 3:00:062017/7/1 3:00:08

122345

101102102103104105

2017/7/1 3:00:10 6 106

2017/7/1 3:00:002017/7/1 3:00:022017/7/1 3:00:062017/7/1 3:00:082017/7/1 3:00:10

12345

101102104105106

(1) (2)

(c)

2017/7/1 3:00:002017/7/1 3:00:022017/7/1 3:00:022017/7/1 3:00:042017/7/1 3:00:062017/7/1 3:00:08

112345

101101102103104105

2017/7/1 3:00:10 6 106

2017/7/1 3:00:002017/7/1 3:00:022017/7/1 3:00:022017/7/1 3:00:042017/7/1 3:00:062017/7/1 3:00:08

112345

101101102103104105

2017/7/1 3:00:10 6 106

(3)


20

Operating procedure

1. Select [Operation] [Data interpolation].

When the [Data interpolation] menu is used, "File was read. (There is an interpolation)" is display in

"LogComment".

Precautions

• Setting for data interpolation is applied as the common setting of Reader. Data interpolation cannot be enabled or disabled

per unit.

• Data is interpolated when "Time of the first line" is set for "Starting condition type" and "Time of the last line" is set for "End

condition type" in a unit definition and the interval between the end time of the previous section and the start time of this

section is longer than the sampling interval. However, if the interval is equal to or longer than 100 sampling intervals, data

interpolation will not performed.


Changing a unit definitionThe following shows the operating procedure for changing the added unit setting.

Operating procedure

1. Right-click on the variable list, and select a target unit in [Select unit definition].

2. Right-click on the variable list, and select [Change unit definition].

For operations for changing the unit setting, refer to the following section.


Precautions

If changing the unit definition of a CSV file after reading the file once, perform the following operations:

When variables of the target unit are being diagnosed in simple diagnosis of Analyzer, stop the diagnosis.

Restart Analyzer and Monitor.

When variables of the target unit is used for the logic of Analyzer, edit the logic as necessary.

When variables of the target unit are used for the simple diagnosis, set the diagnosis rules again.


20

Deleting a unit definitionThe following shows the operating procedure for deleting the unit that is added last.

Operating procedure

1. Right-click on the variable list, and select the unit that is added last in [Select unit definition].

2. Right-click on the variable list, and select [Delete unit definition].


The unit definition is deleted.

The dataset and the response data definition file of the deleted unit are also deleted.

Precautions

• Unit definitions can be deleted in order from one that is added last, but an arbitrary unit definition cannot be deleted.

• To delete an arbitrary unit definition or multiple unit definitions in a batch, refer to the following:

Page 154 Moving/Batch-deleting units


Adding variablesThe following shows the operating procedure for adding the variables of a unit added in the following: (Page 131 Adding

a Unit Definition)

Operating procedure

1. Select [File] [Select unit definition], and select a unit to add its variables.

2. Right-click on the variable list, and select [Add variable].

3. Set each item for a sample CSV file and click the [Next] button.


Page 136 Considerations for setting a sample CSV file

*1 A file in a folder with the encryption attribute enabled cannot be specified.*2 When the name format of a CSV file to be read is 'LOG001_NNNNNNNN.csv', specify it as 'LOG001*.csv'.*3 '*' (asterisk) is used as a wild card indicating any 0 or more characters.

Item Description

Row item title line The row number set to "Row item title line" is displayed.

Sample CSV file*1 Specify the path of a CSV file that includes variables to be added (up to 256 characters).

Reading target CSV file name*2,*3 Specify the name of a CSV file to be read in Reader by using the wild card (*) (up to 256

characters).

Unit name The set unit name is displayed.


20

4. Change the type of data for each row as necessary, and click the [Next] button.*1

*1 For a CSV file output in Real-time Flow Manager, changing the type is not necessary since it is set automatically.

For the considerations on changing the data type, refer to the following:

Page 136 Considerations for setting a type

• By clicking a row header (leftmost column), the type of data on the clicked row can be changed. (It can be

changed only to "Row item title," "Data," or "Unused.")

• For the "Row item title" row, specify the row where the title of each column (variable name) is described.

• Only one row can be specified as "Row item title."

5. Change the data type of added variables as necessary, and click the [Done] button.

For the considerations for changing column items, refer to the following:

Page 136 Considerations for the settings of column items

• By clicking a column header (top row), the data type of variable on the clicked column can be changed.

However, the data type of variables other than added ones cannot be changed.

(Example) When adding 'D16' and 'D20,' the data type of only 'D16' and 'D20' can be changed.

• Only one time stamp column is necessary.

• Up to 5000 column items (variables) can be defined for a CSV file.


6. Click the [Save change] button.


The set unit definition is saved.

Precautions

• To change a unit definition of a CSV file after reading the file, refer to 'Precautions' described in the following:

Page 146 Changing a unit definition

• If variable names set in a target unit are different from the ones set in a specified CSV file, an error message appears and

the variables cannot be added. (The variable names of the time stamp column and unused column do not need to be the

same.)

The following shows examples of CSV files that can be specified and ones that cannot.

(Example) When the following variables are currently set in a target unit

■ Available CSV filesAll the set variables are included and they are aligned the same.

(1) Variables can be added even if the variable name of the time stamp column

is different.

(2) A new variable can be added to the rightmost column if the conditions are

satisfied.

(1) (2)


20

■ Unavailable CSV files • One of the set variables, 'Data4' is missing.

• All the set variables are included but they are aligned differently.

• All the set variables are included but they are aligned differently since 'DataX' is inserted.


Exporting the variable listThe following shows the operating procedure for exporting the variable list when modifying a large part of the list.

Operating procedure

1. Select [File] [Variable list export].

2. Specify the save destination and file name, and click the [OK] button.

The file name is 'UNIT_(unit name)' by default.

3. Edit the contents of the CSV file.

For the path of a file to be exported, up to 255 characters (two characters are used for a double-byte

character) including the extension can be specified.

Importing the variable listThe following shows the operating procedure for importing the CSV file which are exported in 'Exporting the variable list'.

Operating procedure

1. Select [File] [Variable list import].

2. Specify the name of the exported CSV file and click the [OK] button.

For the path of a file to be imported, up to 255 characters (two characters are used for a double-byte


Precautions

• A CSV file that can be imported is only the file that changed the contents of the existing item. In addition, if the number of

variables in a CSV file is different from that of the unit definition of import destination, for example when a variable is added

or deleted, the CSV file cannot be imported.

• If changing the name or type of a variable and importing a CSV file after reading the file once, perform the following

operations:





• A CSV file will not be imported if a variable with the same name is included in the file.


20

Deleting unit dataThe following shows the operating procedure for deleting read data per unit.

Operating procedure

1. Right-click on the variable list, and select a unit to delete data in [Select unit definition].

2. Right-click on the variable list, and select [Delete unit data].

3. As the period for deleting data, select "All" or "Last 1 day (YYMMDD)".

Precautions

When performing this operation, the dataset corresponding to the period for deleting data is also deleted.


Moving/Batch-deleting unitsThe following shows the operating procedures for switching the order of units shown in [Select unit definition] menu and

deleting all units in a batch.

Operating procedure

1. Select [Operation] [Moving and deleting units].

2. Exit Analyzer and Monitor.

3. The current unit definitions are listed in the "Present unit list" in the order of definition.

4. Select units in the order for displaying in the menu and click the [Move] button. Selected units move to the "Unit list after

moving".

5. Leave units to be deleted in "Present unit list". After the setting is completed, click the [OK] button.

The response data definition file of a deleted unit is also deleted at the timing when the "Moving and deleting units" screen is

closed.

Precautions

• When performing this operation, be sure to move at least one unit. (Deleting all units cannot be performed.)

• After performing this operation, restart Real-time Statistic Diagnosis Tool.

• After performing this operation, units the order of which has been switched will be in the following states:

All contents of "LogComment" are deleted.

Logics being diagnosed in Monitor are all stopped. (Perform operations to start the logics as necessary.)


20

Defining an extended sectionDifferent extraction methods can be registered for a CSV file for which section definition is already configured in unit definition.

An extended section is regarded as a single unit in Analyzer and Monitor.

Multiple extended sections can be registered for each unit. If an extended section in which "Specify time" is selected for

"Starting condition type" is defined, only that definition can be registered.

Precautions

There is no limit of registration for the number of extended sections, but the operation guarantee applies up to ten of them.

Operating procedure

1. Right-click on the variable list, select the [Select unit definition] menu, and select a unit definition for which the extended

section is registered.

2. Right-click on the variable list, and select [Extended section definition] [Add section definition].

(1) Unit section (section specified in the unit definition)

(2) Extended section

2017/7/1 3:00:002017/7/1 3:00:022017/7/1 3:00:042017/7/1 3:00:062017/7/1 3:00:082017/7/1 3:00:10

ABCCCC

101102103104105106

(1)(2)


3. Set each item and click the [Save] button.


Page 157 Considerations for setting for "Extended section name"

Page 157 Considerations for setting for "Starting condition type"

Considerations for the setting for "End condition type"

Page 158 Considerations for setting "Only extract in the unit section"

Page 158 Considerations for setting "Cutting out interval"/"Cutting out count"

Page 158 Considerations for setting for "Execute logic by file unit before ending section"

Page 158 Considerations for setting for "Update the section of collected data"

Item Setting content

Extended section name Enter an extended section name (up to 32 characters).

Starting condition type Set the data extraction setting (start row).

When "Specify time" is selected, set the time interval starting at 0:00:00 for extracting one section.

For details on other settings, refer to the following:

Page 137 Section start conditions

Offset Specify the time (0.0000 to 2147483647.0000) from when the start condition is satisfied to the beginning

of the extended time.

Text extraction base time Select the time to extract a text variable value either "After the section starts" or "Before end of section".

Use this setting when a value which is not the first record value is preferred.


Offset Set an offset (-2147483647.0000 to 2147483647.0000) from the timing which is set for "Text extraction

base time".

Set a positive value if "Text extraction base time" is set to "After the section starts" and a negative value if

it is set to "Before end of section".


Unit Select either "sec" (second) or "%" for the offset unit. ("%" is for the section length.)

End condition type Set the data extraction setting (end row).

When "Offset from section head" is selected, specify the time from the start time specified as the start

condition to the end of the extended section in seconds for "Offset". (Only a positive value is allowed.)*1

For details on other settings, refer to the following:

Page 140 Section end conditions

Offset Specify the time (-2147483647.0000 to 2147483647.0000) from when the end condition is satisfied to the

end of the extended time.

Only extract in the unit section Select the checkbox when extracting expanded sections in the section set in the unit definition.

Unselect the checkbox when extracting extended sections without regarding unit section.

Cutting out interval/Cutting out count Set these to extract the extended sections at the specified interval and the number of times, save them for

each unit, and perform a logic.

• The extraction interval can be set between 0 and 86400.

• The number of times of extraction can be set between 0 and 2147483647. When specifying '0', the

extended sections are extracted as many as possible within the section.


20

*1 If "Offset from section head" is selected, the extended section cannot be extracted properly when the start condition is satisfied multiple times before the end condition is satisfied.

*2 The temporary section is not saved as a section but only the expansion section is saved.

■ Performing a logic or a simple diagnosis for each fileThe following shows an image when performing a logic or a simple diagnosis for each file.

Precautions

■ Considerations for setting for "Extended section name" • An extended section is regarded as a single unit in Analyzer or Monitor. To cooperate with Real-time Flow Manager, add

the underscore after the extended section name, Otherwise, cooperation will not be performed properly.

Ex.

When the unit name is 'ABCDEFG':

Define the extended section name as 'ABCDEFG_EXT001'.

■ Considerations for setting for "Starting condition type" • When "Specify time" is selected, the end condition cannot be specified.

• While multiple extended sections can be defined for a single unit, if an extended section in which "Specify time" is selected

for "Starting condition type" is defined, another extended section cannot be defined for that unit. If another extended section

is defined, extended sections may not be extracted correctly.

• When specifying an additional AND condition (start) for the start condition type of an extended section extraction, AND

condition must be satisfied before or at the same time when the start condition is satisfied. Otherwise, the start condition of

the extended section extraction cannot be detected.

Execute logic by file unit before ending section*2 Select the checkbox when extracting extended sections for each CSV file and performing a logic or simple

diagnosis for the temporary section which is from the start of the extended section to the extracted

position.

For the execution image, refer to the following:

Page 157 Performing a logic or a simple diagnosis for each file

Update the section of collected data Select the checkbox when also extracting sections from data which is already read.

While the checkbox is selected, extended sections are extracted from collected data when performing

'Switch to CSV reading mode'.

This item is enable only when saving the extended section definitions.

(1) One CSV file

(2) Unit section (section specified in the unit definition)


(a) A logic or a simple diagnosis is performed for the temporary section from the start of the extended section to the extracted position.

Item Setting content

(1) (1) (1)(1)

(2)

(3)

(a)(a)

(a)(a)


■ Considerations for the setting for "End condition type" • When "Until the start of the next section" is selected, an offset cannot be specified.

• When "Time of the last line", "Text", or "Numeric" is selected, specify a negative value for the offset. (A positive value

cannot be specified.)

• Set the section length to at least one second or longer. Otherwise, the data in the created section cannot be used correctly.

• When specifying an additional AND condition (end) for the end condition type of an extended section extraction, AND

condition must be satisfied before or at the same time when the end condition is satisfied. Otherwise, the end condition of

the extended section extraction cannot be detected.

■ Considerations for setting "Only extract in the unit section" • When selecting the checkbox of "Only extract in the unit section", the sections satisfying the start and end conditions within

the unit section range are only extracted.

■ Considerations for setting "Cutting out interval"/"Cutting out count" • Extraction is not performed outside the extended section.

• When a unit section consists of multiple CSV files and '2' or higher number is specified for the number of times of

extraction, extraction may not be performed with the correct number.

■ Considerations for setting for "Execute logic by file unit before ending section"When any of the following conditions are applied, a logic or a simple diagnosis for the succeeding temporary sections is

terminated.

• Deviation from the diagnosis rule has been detected in logic 'Enhanced trace GB' or 'Variable bound pair GB', or simple

diagnosis 'Guard band'.

Page 320 Enhanced trace GB, Page 324 Variable limit pair GB, Page 215 Guard Band Diagnosis

• A subsection specified in logic 'Extract subsection' or 'L-1-Section extraction' is extracted.

Page 278 Extract subsection, Page 283 L-1-Section extraction

■ Considerations for setting for "Update the section of collected data" • To use a section extracted from read data in Analyzer, it needs to be restarted.

• This setting is not available for an extended section that "Time of the first line" is selected for "Starting condition type" and

"Time of the last line" is specified for "End condition type".


(2) Extraction interval

(a) No extraction (outside the extended section)

(1)

(2) (2) (2) (2) (2) (2)

(a)


20

Recovering unit dataThe following shows the operating procedure for read all backup data of a read CSV file again.

This should be operated when the CSV format (row or column type) or the section start or end condition is changed in the unit

definition.

Operating procedure

1. Right-click on the variable list, and select a unit to be recovered in [Select unit definition].

2. Right-click on the variable list, and select [Unit data recovery].


Unit data recovery is started and unit data is deleted.

4. The log 'Delete unit data End' is output in the log list.

5. Select [File] [Switch to CSV file reading mode].

Re-reading all backup data (CSV file) is started.


Changing a variable nameThe following shows the operating procedure to change a variable name.

Changing a variable name has no affect on the CSV file reading processing.

Operating procedure

1. Select and right-click an item in the "Name" column in the variable list, and select [Edit item].

2. Enter a name and click the [OK] button.

Precautions

• If changing the name of a variable included in a CSV file after reading the file once, perform the following operations:





• A variable name cannot be changed to one duplicated with another variable.

• The following characters cannot be used in a variable name:

\ (0x5C), / (0x2F), : (0x3A), * (0x2A), ? (0x3F), " (0x22), < (0x3C), > (0x3E), | (0x7C)

￥ (0x818f), ／ (0x815e), ： (0x8146), ＊ (0x8196), ？ (0x8148), ” (0x8168), ＜ (0x8183), ＞ (0x8184), ｜ (0x8162)

Changing a data typeThe following shows the operating procedure to change the data type of a variable.

Operating procedure

1. Select and right-click an item in the "Type (numerical value | char)" column in the variable list, and select [String],

[Numeric], or [varn] in [Edit item].

Precautions

If changing the type of a variable included in a CSV file after reading the file once, perform the following operations:






20

Entering an operational expressionBy entering an operational expression, the result can be used as data to be used in Analyzer.

Variable values are represented as 'x'. (Use such as 'x+100')

Symbols that can be used for an operation expression are *, /, +, -, (, and ).

Operating procedure

1. Select and right-click an item in the "Expression" column in the variable list, and select [Edit item].

2. Enter an operational expression and click the [OK] button.

16220 Reader20.6 Switching the Operation Mode to CSV File Reading Mode

20.6 Switching the Operation Mode to CSV File Reading Mode

When a CSV file is stored in 'CSV input folder', Reader automatically switches to the CSV file reading mode.

Operating procedure

Select [File] [Switch to CSV file reading mode].

• To edit a unit definition, select [File] [Switch to setting edit mode].

• When a unit definition has been added, changed, or deleted, restart Analyzer and Monitor.

• By performing the following procedure when starting Real-time Statistic Diagnosis Tool, Reader will start in

the CSV file reading mode, however, if any of the paths set for "Reader equipment data folder", "CSV input

folder", and "RSD dataset folder" is not defined, Reader will start in the setting edit mode.

Select [Mitsubishi Electric Edgecross Software] [Real-time Data Analyzer] from Windows Start, then

right-click [Real-time Statistic Diagnosis Tool] and select [Properties] from the shortcut menu.

Add ' /R' at the end of the path which is set for "Target" in the property screen.

20 Reader20.6 Switching the Operation Mode to CSV File Reading Mode 163

20

Precautions

• If the CSV input folder contains a read-only file, Reader cannot read a CSV file.

• When a CSV file which contains data which duplicates data in an already read CSV file or data with the past time stamp is

read, this CSV file is regarded to be invalid. Be careful if CSV files can be output after reversing the time. In this case, the

invalidated CSV files are stored in the following folder under the Reader equipment data folder.

Reader_UNIT\YYMMDD\Original\Invalid ('YYMMDD' indicates the current date.)

• Do not store a file with a name not corresponding to a CSV file name specified in the unit definition to the 'UNIT' folder

under 'CSV input folder'. Otherwise, Reader loads the file as an invalid file in the CSV file reading mode and saves it in the

following folder under 'Reader equipment data folder'.

Reader_UNIT\YYMMDD\Original\Invalid


• Real-time Statistic Diagnosis Tool does not support the daylight saving time. A CSV file containing a time stamp at the end

of the daylight saving time may not be read in Reader.

• A CSV file to be loaded into the CSV input folder must have two or more data rows.

• A CSV file to be loaded into the CSV input folder must be the file that the period between the first and last data rows is 10

seconds or longer. If CSV files that the period is shorter than 10 seconds are loaded consecutively, the file loading

processing may be delayed.

• When reading a variable of which "Type (numerical value | char)" is "String", the first and last single-byte spaces of the

variable value are deleted. (Double-byte spaces are not deleted.)

• When the size of each CSV file to be loaded into the 'CSV input folder' is too large, a memory overflow error occurs and the

file may not be read. In this case, reduce the size of each file by reducing the number of variables or that of records in the

file. (It is recommended to reduce the file size to 10 M bytes or less.)

• For switching the mode to the CSV file reading mode, the paths set for all "Reader equipment data folder", "CSV input

folder", and "RSD dataset folder" need to be defined.

• For reading a CSV file, start Monitor in advance. If a CSV file which is stored in the specified folder cannot be read while the

mode is switched to the CSV file reading mode, check whether Monitor is running. (Page 416 Monitor)

16421 Analyzer

21 Analyzer

This software analyzes and diagnoses data in a CSV file.

The following shows the procedure for using Analyzer.

Operating procedure

1. Set the items in the 'Data management setting'.

Page 390 Data Management Setting

2. Analyze data.

• Display and analyze a waveform of read data


• Display and analyze statistics of read data


• Display and analyze frequency of read data


• Create a correlation matrix and perform advanced analysis


Page 243 Multiple Regression Analysis (LMR)

Page 250 Maharanobis - Taguchi method (MT)

3. When displaying and analyzing a data waveform, a single waveform can be analyzed in detail by configuring a data set.

Page 384 Management

Page 365 Displaying Data after Running

4. Create the diagnosis rule and start diagnosis.

• Apply the SPC rule to statistic and run diagnosis

Page 189 SPC

• Perform a diagnosis in multivariate analysis (single central value format).


• Set a guard band on the waveform and run diagnosis


• Program a diagnostic logic and run diagnosis


21 Analyzer21.1 Displaying Read Data 165

21

21.1 Displaying Read DataShows CSV file data read by Reader as a waveform.

Waveforms of data from multiple sections can be overlapped or arranged horizontally for display or analysis.

When arranged horizontally, each section is shown as a single waveform and sections are not consecutive.

Operating procedure

1. Select [Disp] [Read data].

2. Select data to read.

Page 179 Selecting read data

3. Change the display and perform analysis according to the screen structure.

Page 166 Screen configuration

16621 Analyzer21.1 Displaying Read Data

Screen configurationThe following shows the screen configuration for displaying read data.

Displayed items

Item Description

(1) Trend graph (waveform) display screen The trend graph (waveform) of a variable selected in "Select variable" is displayed.

• The vertical axis is automatically scaled according to the maximum and minimum values of the

displayed waveform.

• When the [Overlay] button is selected, the start time and elapsed time of a single section is displayed

in seconds on the horizontal axis. In other cases, the time is shown.

The display status of the waveform currently snapped (grabbed) by the cursor can be switched to hide.

Page 170 Hiding a waveform

Equipment The fixed name "UNIT" is shown.

When nothing is displayed, right-click and select 'RSD dataset folder' set in Reader to display "UNIT".

Select unit Select the unit name or expansion section name to display data.

If multiple units (including expansion sections) are defined in the Reader settings, right-click and select

other units.

When starting Analyzer, the unit which is added last in Reader is displayed by default.

From/To Specify a section including data to be displayed in the section list.

Page 170 Setting "From/To"

The section list is shown by clicking the [Sctn] button.

Page 172 Selecting section list

[Filter text name specification] button Select a text variable to filter data to view.

Text variables are shown when the read CSV file contains text variables.

Specify filtering text Specify a value of a text variable selected by clicking the [Filter text name specification] button (up to 256

characters).

All data for sections for the specified text variable value is shown.

• When "*" is specified, all data for sections for which the selected text variable is not null is shown.

• When "(null)" is specified, all data for sections for which the selected text variable is null is shown.

• When nothing is specified, data for all sections is shown. (Data is not filtered.)

Histogram*1 When the checkbox is selected, the histogram of the data in the display section is displayed.

When multiple variables are selected, they are displayed in different colors. (The values on the X and Y

axes and the cursor are hidden.)

[Var.] tab*2 Select variable Select a variable to be displayed for a specified equipment, unit, and period.

Multiple variables can be selected and their waveforms can be overlapped.

[EQ Def] tab*3 Shows instrumental errors.

This function compares variables with the same name between different units.

Data from different units can be displayed simultaneously to overlap their waveforms and compare their

statistics.

[Display variable] button Select a variable to display the instrumental error.

Equipment "UNIT" is displayed.

Unit Select the name(s) of unit(s) to view data simultaneously.

(1)

(2)

(3)


21

[Sctn] button Click this to display the section list.

Select data to display in the trend graph (waveform) display screen for each section read by Reader.


[Save] button Saves information (of which section of which variable) concerning the display data.

If the file is saved in a folder other than 'Reader equipment data folder' and 'RSD dataset folder, the file

needs to be moved manually to the migration destination when migrating the settings to Real-time

Statistic Diagnosis Tool installed in another industrial PC.

Page 121 Migrating settings to Real-time Statistic Diagnosis Tool in another industrial PC, Page 122

Project File Setting Utility

[Read] button Reads and displays information concerning saved data (of which section of which variable).

[Multi] button Shows multiple (up to six) variables in separate windows arranged on the screen.

Click again to return to the previous display.

[Overlay] button Overlays data of the selected variables for each section read by Reader.

Data for multiple sections as well as multiple data records can be overlaid.


[Expand] button*4 Expands a range specified by dragging on the display screen.

The same operation as clicking is performed. (By clicking this button, is selected.)

[V expand] button*4 Expands a range specified by dragging on the display screen vertically without changing the horizontal

scale.


[H expand] button*4 Expands a range specified by dragging on the display screen horizontally without changing the vertical

scale.


[Zoom in] button*4 Expands the display around a point clicked on the waveform display screen, gradually as the point is

clicked.


[Zoom out] button*4 Reduces the display around a point clicked on the waveform display screen, gradually as the point is

clicked.


[Drag] button*4 Drags to move the waveform on the display screen.


When selection is canceled and the cursor is displayed, it can be used to select a waveform.

For showing or hiding the cursor, refer to the following:

Page 178 Show and Hide Cursor

[Reset] button Resets the effect of the following buttons and returns the display to the initial state.

• [Expand] button,

• [V expand] button,

• [H expand] button,

• [Zoom in] button,

• [Zoom out] button,

• [Drag] button,


[Fix axis] button When displaying a waveform, disables automatic adjustment of the upper and lower limit values of the

axis and fixes the axis display.

Clicking this button again cancels selection.

[Lock] button*1 Locks the displayed waveform and retains its display even when a new variable is selected.

Clicking this button again cancels selection.

[Scl-fit.] button*1 Expands each of displayed waveforms so that its amplitude fits the vertical axis range of the area.

The values of the vertical axis are aligned with the variable listed top in the "Select variable" list.

Amplitudes are aligned for the display purpose only and the data itself is not changed. When output to a

CSV file, the original data is output.

[Max] button*3,*5 Calculates the maximum value for each waveform per section, and displays a waveform connecting the

maximum values.

When the [Overlay] button is selected, the maximum value of the overlapped waveforms is calculated for

each step on the horizontal axis, and a waveform connecting the calculated values is shown.

[Ave] button*3,*5 Calculates the average value for each waveform per section, and displays a waveform connecting the

average values.

When the [Overlay] button is selected, the average value of the overlapped waveforms is calculated for


[Min] button*3,*5 Calculates the minimum value for each waveform per section, and displays a waveform connecting the

minimum values.

When the [Overlay] button is selected, the minimum value of the overlapped waveforms is calculated for


Item Description


[Range] button*3,*5 Calculates the maximum and minimum values for each waveform per section, and displays waveforms

connecting the maximum values and connecting the minimum values.

When the [Overlay] button is selected, the difference of the maximum and minimum values of the

overlapped waveforms is calculated for each step on the horizontal axis, and a waveform connecting the

calculated values is shown.

[Center] button*3,*5 Calculates the median for each waveform per section, and displays a waveform connecting the

medians.

When the [Overlay] button is selected, the median value of the overlapped waveforms is calculated for


[Sigma] button*3,*5 Calculates a sigma value for each waveform per section, and displays a waveform connecting the sigma

values.

When the [Overlay] button is selected, the sigma value of the overlapped waveforms is calculated for


[Parallel] button When data in multiple sections is displayed, shifts each unit of data vertically to show them individually.

When multiple variables are selected, they are displayed in different colors.

In the parallel display, values on Y axis and the cursor will not be displayed.

[3D] button*6 Draws a 3D chart for multiple sections of a variable.

Page 174 Drawing a three-dimensional graph

[CSV for matrix] button Outputs all data shown to a CSV file in the format that can be read by 'Create correlation matrix'.


A save destination folder for a CSV file to output can be specified arbitrarily. (For the path of a file to be

output, up to 255 characters (two characters are used for a double-byte character) including the

extension can be specified.)

However, do not specify a folder with the encryption attribute enabled.

If the file is saved in a folder other than 'Reader equipment data folder' and 'RSD dataset folder, the file

needs to be moved manually to the migration destination when migrating the settings to Real-time

Statistic Diagnosis Tool installed in another industrial PC.

Page 121 Migrating settings to Real-time Statistic Diagnosis Tool in another industrial PC, Page 122

Project File Setting Utility

When data in multiple sections is displayed in the screen of 'Read data display,' data in each section is

output as a distinct variable even if there is only one variable in the whole data. (It is output as '[variable

name]_XX.' A sequential number is assigned to 'XX' per section.)

[CV] button Applies 'MA' and/or 'difference' conversion to the vertical axis of the displayed data.

Page 175 Conversion

[FLT] button Sections a waveform according to changes of the value of the specified variable.

Page 176 Extracting waveforms

[Update] button*7 Updates section data to display in the section list to data of the latest date of a loaded CSV file. (The

date specified for "From" and "To" are updated.)

If filtering the section list, the filtering will be reset. However, hiding of the section list is not reset.


Up/Down/Left/Right button Clicking the left or right button moves the cursor on the chart to the left or right, respectively, by one

point.

The up and down buttons cannot be used.

These buttons are effective while the cursor is shown on the chart.



When the cursor is displayed, it can be used to select a waveform.



Item Description


21

*1 Cannot be checked while the [Multi] button is selected.*2 Not shown if instrumental error display is disabled and only the "Select variable" items are shown.

Page 390 Data Management Setting*3 It is shown instrumental error display is enabled in the 'Data management setting'.

Page 390 Data Management Setting*4 One of the [Expand], [V expand], [H expand], [Zoom in], [Zoom out], and [Drag] buttons can be selected. Clicking the same button again

cancels its effect.*5 Only one of the following buttons can be selected simultaneously: [Max], [Ave], [Min], [Range], [Center], and [Sigma] buttons. Clicking

the same button again cancels its effect.*6 Clickable when the [Overlay] button is selected.*7 The unit name list displayed in "Select unit" is not updated. When updating the list after adding or deleting a unit definition in Reader,

restart Analyzer.

Expands a range specified by dragging on the display screen.

The same operation as clicking the [Expand] button is performed. (By clicking this button, the [Expand]

button is selected.)

Expands a range specified by dragging on the display screen vertically without changing the horizontal

scale.

The same operation as clicking the [V expand] button is performed. (By clicking this button, the [V

expand] button is selected.)

Expands a range specified by dragging on the display screen horizontally without changing the vertical

scale.

The same operation as clicking the [H expand] button is performed. (By clicking this button, the [H

expand] button is selected.)

Resets the effect of the following buttons and returns the display to the initial state.

• [Expand] button,

• [V expand] button,

• [H expand] button,

• [Zoom in] button,

• [Zoom out] button,

• [Drag] button,

The same operation as clicking the [Reset] button is performed. (By clicking this button, the [Reset]


Expands the display around a point clicked on the waveform display screen, gradually as the point is

clicked.

The same operation as clicking the [Zoom in] button is performed. (By clicking this button, the [Zoom in]


Reduces the display around a point clicked on the waveform display screen, gradually as the point is

clicked.

The same operation as clicking the [Zoom out] button is performed. (By clicking this button, the [Zoom

out] button is selected.)

Drags to move the waveform on the display screen.

The same operation as clicking the [Drag] button is performed. (By clicking this button, the [Drag] button

is selected.)

When selection is canceled and the cursor is displayed, it can be used to select a waveform.



(2) Cursor legend The legend of the waveform currently snapped (grabbed) by the cursor can be displayed.

Effective while the cursor is shown on the chart.



(3) Chart legend Shows the legend of the waveform chart.

When multiple variables are selected, they are displayed in order from the variable listed top in the

"Select variable" list.

The chart can be shown or hidden by checking or unchecking the legend.

The name of a legend can be changed arbitrarily, but it returns to the original name if selecting a variable

again.

Item Description


Hiding a waveformThe display status of the waveform snapped (grabbed) by the cursor can be switched to hide.

Operating procedure

1. Right-click in the display screen and select [Clear the waves.] from the shortcut menu.

When redisplaying a waveform, select the section corresponding to the hidden waveform in the section list.

(When hiding a waveform, the corresponding section will be unselected in the section list.)

Selecting section list

Setting "From/To"Specify a section including data to be displayed in the section list.

The section list is displayed by clicking the [Sctn] button.


Precautions

Section data newly read by Reader is not automatically displayed in Analyzer. To display section data newly read by Reader in

Analyzer, set the date in "To" again or click the [Update] button.

■ Setting "From"Specify the start date of the section data to be displayed in the section list in the format 'HH:MM:SS.sss YYYY/mm/dd'.

The reference time*1 of the latest date of data in a unit selected for "Select unit" is set by default.

*1 Time which is set for "Default retrieval period" (Page 131 Adding a Unit Definition)

Operating procedure

1. Click .

2. Specify the time and date. After selecting units, click the [OK] button.

Clicking [Set Time to Now] sets the time to the current time.

• 'YYYY/mm/dd' cannot be set by entering values directly.

• Clicking the [Cancel] button discards the setting and closes the "Set Time and Date" screen.

• In each field of "From" and "To", the year, month, and date are displayed. (On the tooltip, the hour, minute,

and second are also displayed.)


21

■ Setting "To"Specify the end date of the section data to be displayed in the section list in the format 'HH:MM:SS.sss YYYY/mm/dd'.

The date and time calculated by adding the time which is set for "Default retrieval period" in 'Add unit definition' of Reader to

the time which is set for "From" is set by default.


Operating procedure

1. Click .

2. Specify the time and date. After selecting units, click the [OK] button.

Clicking [Set Time to Now] sets the time to the current time.

• 'YYYY/mm/dd' cannot be set by entering values directly.

• When changing a unit in "Select unit", the date and time which is set for "To" is changed according to the

setting for "Default retrieval period" of the changed unit.

• Clicking the [Cancel] button discards the setting and closes the "Set Time and Date" screen.

• In each field of "From" and "To", the year, month, and date are displayed. (On the tooltip, the hour, minute,

and second are also displayed.)


Selecting section listClick this to display the section list.

Select data to display in the trend graph (waveform) display screen for each section read by Reader.

The following explains the menu items available by right-clicking on the section list:

Page 172 Menu items available by right-clicking on the section list

Operating procedure

1. Click the [Sctn] button.

2. Select a section in the section list. Data of a selected section is displayed in the trend graph (waveform) display screen.

Data of multiple sections can also be selected.

■ Menu items available by right-clicking on the section list

The deleted section data can be restored by deleting "Reader_UNIT\(unit number)_(unit

name)\BlockMask.bin" in 'Reader equipment data folder' set in Reader and then restarting Analyzer or

choosing [Reset Hiding].


Item Description

Narrow Down (column item name) Filters section data shown in the section list.

Data is filtered by the selected text variable value.

Narrow DownReset Cancels filtering of the section list and returns it to the state before filtering.

Hide Temporarily hides section data shown in the section list.

Highlight part Hides the highlighted section data.

Other than highlight Hides the section data not highlighted.

HideReset Cancels hiding of the section list and returns it to the state before hiding.

Out-CSV Outputs the section list to a CSV file.

When selected, the "Out-CSV" screen appears.

Page 173 "Out-CSV" screen

Delete section Deletes the highlighted section data from the section list.

Description and Tip Do not use this.


21

■ "Out-CSV" screen

Window

Right-click on the section list and choose [Out-CSV].

Displayed items

Item Description

Comment Insert a text entered in this item in the first row of a CSV file to be saved.

"^CSV" is inserted on the line following the comment to indicated the end of the comment.

Only highlight lines If checked, only the highlighted lines are output to the CSV file.

Delimiter • Comma: Saves in the CSV format.

• Tab: Saves in the text format.

[Save] button Saves the CSV file by specifying the location and file name.

Do not specify a folder with the encryption attribute enabled as a save destination.

[Cancel] button Discards the setting and closes the "Out-CSV" screen.


Drawing a three-dimensional graphDraw three-dimensional graph of multiple sections for one variable.

When multiple variables are selected, the value of the variable listed top in the "Select variable" list is shown.

Operating procedure

1. Click the [3D] button.

2. Shows the "3D contour line" screen.

The view point of a drawn three- dimensional graph can be changed by dragging the graph.

The X-axis, Y-axis, and Z-axis represents the section data, time, and variable value, respectively.

3. Switch the graph display by using the following buttons as necessary.

• Data of up to 100 sections is shown.

• The values on Z axis is displayed in different colors.

Item Description

Displays a graph with X and Y axes.

Displays a graph with X and Z axes.

Displays a graph with Y and Z axes.

Displays a graph with X, Y, and Z axes.


21

ConversionApply 'MA' and/or 'Difference' conversion to the vertical axis of the displayed data.

Only the odd number of average calculation target points can be specified for moving average conversion.

Operating procedure

1. Click the [CV] button.

2. Set each item.

The vertical axis of the displayed waveform is converted.

*1 Only odd numbers can be specified.

Item Description

MA When checked, the moving average of up to the current point for the specified points is calculated and shown as the

value of the current point.

If the current point falls below the specified number of points, the original value is shown.

Num of points Specify the number of points (3 to 2147483647)*1 to calculate the moving average.

Difference When checked, the difference between the values of the current point and the point before it for the number of

specified points is calculated and shown as the value of the current point.

If no value exists at the point before the current point for the number of specified points, "NaN" is shown.

Num of Points Specify the number of points (-2147483648 to 2147483647) to calculate the difference.


Extracting waveformsSections a waveform according to changes of the value of the specified variable.

A waveform can be extracted by specifying a step variable or waveform.

Waveform display reverts to the original state by closing the "Filter display range by variable" screen or

clicking the [FLT] button again.

■ Specifying step variable

Operating procedure

1. Click the [FLT] button.

2. Select a step variable and set each item.

The waveform of the selected variable is shown.

A waveform of the topmost section data selected in the section list is displayed.


3. Click the [Update screen] button.

The part of the waveform with a black background is extracted and displayed in the trend graph (waveform) display screen.

Precautions

• If the start condition is not satisfied, the waveform is not extracted.

• If the end condition is not satisfied, the waveform is extracted to the end of the section.

Item Description

Start condition Select a step variable value (-10000.0000 to 10000.0000) for the start condition.

The point where the step variable changes to the specified value is the start point of waveform extraction.

Start offset (sec) Specify an offset from the start condition in seconds (-10000.0000 to 10000.0000).

When offsetting backward, specify a negative value. When offsetting forward, specify a positive value.

End condition Select a step variable value (-10000.0000 to 10000.0000) for the end condition.

The point where the step variable changes to the specified value is the end point of waveform extraction.

End offset (sec) Specify an offset from the end condition in seconds (-10000.0000 to 10000.0000).



21

■ Specifying waveform

Operating procedure

1. Click the [FLT] button.

2. Select a variable other than a step variable and set each item.

The waveform of the selected variable is shown.

A waveform of the topmost section data selected in the section list is displayed.


3. Click the [Update screen] button.

The part of the waveform with a black background is extracted and displayed in the trend graph (waveform) display screen.

Precautions

• If the start condition is not satisfied, the waveform is not extracted.

• If the end condition is not satisfied, the waveform is extracted to the end of the section.

Item Description

Start condition Enter a variable value (-10000.0000 to 10000.0000) to be the start condition and select its change type from the

following items.

• UP: Value changes from less than the input value to equal to or greater than it.

• DOWN: Value changes from greater than the input value to equal to or less than it.

The point where the step variable changes to the specified value is the start point of waveform extraction.

Start offset (sec) Specify an offset from the start condition in seconds (-10000.0000 to 10000.0000).


End condition Enter a variable value (-10000.0000 to 10000.0000) to be the end condition and select its change type from the

following items.

• UP: Value changes from less than the input value to equal to or greater than it.

• DOWN: Value changes from greater than the input value to equal to or less than it.

The point where the step variable changes to the specified value is the end point of waveform extraction.

End offset (sec) Specify an offset from the end condition in seconds (-10000.0000 to 10000.0000).



Show and Hide CursorShows and hides the cursor.

The second cursor can also be shown as well.

Operating procedure

1. Right-click on the display screen, and select an item from [Cursor].

2. To show the second cursor, right-click on the 'cursor legend', and select [Create Cursor] [Single-Plot].

3. Right-click on the 'cursor legend' and select [Display item] [Vertical Scrollbar].

The scroll bar is shown in the 'cursor legend'.

• When two cursors are displayed, dragging them can move the cursor position.

• When a cursor is displayed, the time and value of the location on which a waveform is snapped are

displayed on the upper left of the trend graph (waveform) display screen.

Item Description

+ Shows the cross-hair cursor.

| Shows the vertical-bar cursor.

-- Shows the horizontal-bar cursor.

Delete Hides all cursors shown.

Displayed label Do not use this.


21

Selecting read data Select data to read.

Operating procedure

1. When "UNIT" is not displayed for "Equipment", right-click on the display area of "Equipment".

2. Select the RSD dataset folder set in Reader.

3. "UNIT" is displayed in "Equipment".

4. Select a unit name to display data or an extended section name for "Select unit".

5. Select a variable to display from "Select variable".

"UNIT" is displayed in "Equipment" by satisfying all of the following conditions: A unit name or extended

section name can be selected for "Select unit."

Folders are defined in Reader.


The operation for switching the operation mode of Reader to the CSV file reading mode has been

performed after the operations in is completed.


Analyzer is restarted or the [Update] button is clicked after the operation in is completed.

If a folder defined in Reader is changed, perform the operations in and again; otherwise, data of the

previous folder may be displayed.

18021 Analyzer21.2 Displaying Read Data Statistics

21.2 Displaying Read Data StatisticsCalculates statistic for each variable of data in a CSV file, which was read by Reader, in each section, and display the result

by linking the period specified for "From" and "To".

By arranging statistics data for multiple variables in a single window or multiple windows for analysis, changes in statistics can

be understood.

Operating procedure

1. Select [Disp] [Read data statistics].





21 Analyzer21.2 Displaying Read Data Statistics 181

21

Screen configurationThe following shows the screen configuration for displaying read data statistics.

Displayed items

Item Description

(1) Trend chart (statistics) display screen Shows the statistics for a variable selected from the "Select variable" list.

• The vertical axis is automatically scaled according to the maximum and minimum values of the

displayed waveform. Note that when multiple variables are selected, the vertical axis values are not

displayed.

• Numbers corresponding to the number of units of data (sections) included in the period specified for

"From" and "To" are displayed on the horizontal axis.

[Var.] tab*1 Select variable Select the statistic of a variable to be displayed for a specified equipment, unit, and period.

By selecting multiple variables, the statistics can be displayed vertically.


The statistics of all sections are displayed in the statistic display.

When selecting the section data in the section list while the cursor is displayed, the waveform of the

section is snapped by the cursor. In addition, when snapping a waveform by the cursor, the

corresponding section data in the section list is selected.

When hiding the data of all sections in the section list, the statistics of all sections before narrowing down

or hiding is displayed.


[Multi] button Shows the statistics of multiple (up to six) variables in separate windows arranged on the screen.


[Difference] button Shows a differential value of two successive points on a waveform.

'Y value of a succeeding point' - 'Y value of a preceding point' is shown.

If the Y value is increasing, a positive value is shown.

[Max] button*2 Click this to display the maximum value of a waveform for each section.

[Ave] button*2 Click this to display the average value of a waveform for each section.

[Min] button*2 Click this to display the minimum value of a waveform for each section.

[Range] button*2 Click this to display the difference between the maximum and minimum values of a waveform for each

section.

[Center] button*2 Click this to display the median of a waveform for each section.

[Sigma] button*2 Click this to display the standard deviation of a waveform for each section.

[Integration] button*2 Click this to display the integral value (area) for each section.

[Length] button*2 Shows the number seconds from the start point of the waveform per section.



For how to show and hide the cursor and cursor legend, refer to the following:


(1)

(2)

(3)

18221 Analyzer21.2 Displaying Read Data Statistics

*1 Not shown if instrumental error display is disabled and only the "Select variable" items are shown.Page 390 Data Management Setting

*2 Only one of the following buttons can be selected: [Max], [Ave], [Min], [Range], [Center], [Sigma], [Integration], and [Length] buttons.

The other items are the same as those described in the following section.


Note that the following items are not included.

• [Scl-fit.] button

• [Overlay] button

• [Parallel] button

• [3D] button

• [CV] button


When multiple variables are selected, they are displayed in order from the variable listed top in the

"Select variable" list.


The name of a legend can be changed arbitrarily, but it returns to the original name if selecting a variable

again.

Item Description

21 Analyzer21.3 Displaying Read Data Frequency 183

21

21.3 Displaying Read Data FrequencyCalculates the moving average for the CSV file data read by Reader and displays the result after removing noises.

The frequency filter can also be applied.

Further, a spectrogram after an STFT transformation or a scalogram after a Wavelet transform can be displayed.

Operating procedure

1. Select [Disp] [Read data frequency].


Page 188 Selecting read data (frequency display)



18421 Analyzer21.3 Displaying Read Data Frequency

Screen configurationThe following shows the screen configuration for displaying read data frequency.

Window

Displayed items

■ Screen items

Item Description

(1) Trend graph (waveform) display screen Shows the waveform for a variable selected by the [Display variable] button.

The display status of the waveform currently snapped (grabbed) by the cursor

can be switched to hide.


(2) Spectrogram/scalogram display screen Shows the spectrogram/scalogram for a variable selected by the [Display

variable] button.

(3) Power spectrum/energy display screen Shows the power spectrum/energy for a variable selected by the [Display

variable] button.

[Display variable] button Select the frequency of a variable to be displayed for a specified equipment, unit,

and period.

[3D] button Draws a three-dimensional graph of variables for one section.

Page 187 Drawing a three-dimensional graph

(3)

(2)

(1)

(5)

(4)


21

[Filter] tab Configures settings related to filtering.

MA (noise removal) Performs smoothing using the moving average.

Frequency filter • When "Low pass" or "High pass" is selected, specify a value for "Cut-off

frequency 1[Low level](Hz)".

• When "Band pass" or "Band stop" is selected, specify values for "Cut-off

frequency 1[Low level](Hz)" and "Cut-off frequency 2[High level](Hz)".

However, a value specified for "Cut-off frequency 1[Low level](Hz)" must be

smaller than one specified for "Cut-off frequency 2[High level](Hz)".

Cut-off frequency 1[Low level](Hz) Set the low cut-off frequency (0.0000 to 100000.0000).

Cut-off frequency 2[High level](Hz) Set the high cut-off frequency (0.0000 to 100000.0000).

[IIR filter] tab Configures settings related to the IIR (Infinite Impulse Response) filter.

Topology Select a topology.

Degree Set a degree (0 to 100).

[FIR filter] tab Configures settings related to the FIR (Finite Impulse Response) filter.

Tap number Set the number of taps (0 to 1000).

Head offset Set the offset (number of points) to eliminate influence by the leading part of a

section.

The range of available values is 0 to 1000.

The part of the waveform from the lead for the specified offset is inverted and

connected to the preceding part.

After applying the filter, the lead part of the section for the specified offset is cut

off.

Error It is lighted in red if an error occurs in the filter setting*1. (It is lighted in green if no

error occurs.)

[Conversion type] tab Configures settings related to the conversion type.

STFT Configures settings related to STFT.

Number of shift points Set the number of points to switch the sliding window (1 to 1000).

By shortening the time step, the time resolution improves but the frequency

resolution becomes worse.

Frequency bin Set the FFT size (number of points) for STFT.

The range of available values is 2 to 512.

By increasing the FFT size, the frequency resolution improves.

Window type Enter the window information used to calculate STFT.

Window parameter • Kaiser window: parameter

• Dolph-Chebyshev window: Proportion s between main robe and side robe

• Gauss window: Standard deviation

• Other types: Entry is ignored.

The default for the window parameter is NaN.

The range of available values is '0.000' to '10000.0000'.

Window length Set the window length (0 to 10000).

Wavelet Configures settings related to Wavelet.

Y axis scale Select "Frequency" or "Scale".

• "Frequency"

The vertical axis represents the frequency scale from '0' to 'sampling interval/2'.

It is a reverse display to the wavelet scale.

• "Scale"

The consecutive wavelet calculation is performed using a positive scale.

The vertical axis represents the wavelet scale.

Scale Set the number of scales used in wavelet calculation (1 to 10000).

• The frequency resolution improves for larger scales.

• The time resolution improves for smaller scales.

Number of shift points Set the number of seconds for shift and conversion (1 to 10000).

Calculates the wavelet coefficient per shift. (Discrete wavelet transform)

Range-scaling Select "Energy" or "Amplitude".

• "Energy"

Energy remains the same for all scales.

• "Amplitude"

Each scale has the maximum value of the frequency response values.

Selection of color Selects the color type for the spectrogram/scalogram display screen.

Item Description


*1 If an error occurs, a waveform when setting "None" for "Frequency filter" will be displayed.*2 The cursor can also be displayed in the trend graph (waveform) display screen. (It cannot be displayed in the power spectrum/energy

display screen.)




• [Multi] button


• [CV] button

• [FLT] button

• [Fix axis] button

• [Lock] button


• [Max] button

• [Min] button

• [Ave] button

• [Range] button

• [Center] button

• [Sigma] button


• Histogram

• [Var.] tab

• [EQ Def] tab

(4) Cursor legend The legend of the waveform currently snapped (grabbed) by the cursor in the

spectrogram/scalogram display screen can be displayed.

It is available while the cursor is displayed in the spectrogram/scalogram display

screen.

For showing or hiding the cursor, refer to the following:*2




The name of a legend can be changed arbitrarily, but it returns to the original

name if selecting a variable again.

Item Description


21

Drawing a three-dimensional graphDraw a three-dimensional graph for one section of a variable.

When multiple sections are selected, a graph for the section with the smallest number is displayed.

Operating procedure

1. Click the [3D] button.

2. The "3D strength graph" screen appears.

The view point of a drawn three- dimensional graph can be changed by dragging the graph.

The X axis, Y axis, and Z axis indicate time, frequency, and power spectrum (when the conversion type is 'STFT') or energy

(when the conversion type is 'Wavelet') respectively.

3. Switch the graph display by using the following buttons as necessary.

The values on Z axis is displayed in different colors.

Item Description

Displays a graph with X and Y axes.

Displays a graph with X and Z axes.

Displays a graph with Y and Z axes.

Displays a graph with X, Y, and Z axes.


Selecting read data (frequency display) Select data to read.

Operating procedure

1. When "UNIT" is not displayed for "Equipment", right-click on the display area of "Equipment".

2. Select the RSD dataset folder set in Reader.

3. "UNIT" is displayed in "Equipment".

4. Select a unit name to display data or an extended section name for "Select unit".

5. Click the [Display variable] button and select a variable.

"UNIT" is displayed in "Equipment" by satisfying all of the following conditions: A unit name or extended

section name can be selected for "Select unit."






Analyzer is restarted or the [Update] button is clicked after the operation in is completed.

If a folder defined in Reader is changed, perform the operations in and again; otherwise, data of the

previous folder may be displayed.

21 Analyzer21.4 SPC 189

21

21.4 SPCCalculates statistics of the CSV file data read by Reader per section, and diagnoses the result using the SPC rule.

SPC is a method for visually diagnosing the statistic change of manufacturing data by an abnormality determination rule (SPC

rules) defined by JIS.

This method allows for detecting a sign of a failure in advance during the manufacturing process.

Operating procedure

1. Select [Simple analysis - Simple diagnosis] [SPC].



3. Select an SPC rule used for diagnosis.

Page 192 Selecting SPC rule

4. Create the SPC standard value.

Page 198 Creating SPC

5. Start diagnosis using the SPC rule.

Page 202 Starting diagnosis using SPC rule

6. If the diagnosis rule is changed after starting diagnosis, the changed rule is used.

Page 203 Updating diagnosis definition

7. Stops diagnosis.

Page 203 Stopping diagnosis

19021 Analyzer21.4 SPC

Screen configurationThe following shows the screen configuration for an SPC diagnosis.

Displayed items

Item Description

(1) Trend chart (statistics) display

screen

Shows the statistics for a variable selected from the "Select variable" list.

• The vertical axis is automatically scaled according to the maximum and minimum values of the displayed waveform.

Note that when multiple variables are selected, the vertical axis values are not displayed.

• Numbers corresponding to the number of units of data (sections) included in the period specified for "From" and "To"

are displayed on the horizontal axis.



When selecting the section data in the section list while the cursor is displayed, the waveform of the section is snapped

by the cursor. In addition, when snapping a waveform by the cursor, the corresponding section data in the section list is

selected.

When hiding the data of all sections in the section list, the statistics of all sections before narrowing down or hiding is

displayed.








[Range] button*1 Click this to display the difference between the maximum and minimum values of a waveform for each section.




[Length] button*1 Shows the number seconds from the start point of the waveform per section.

Select variable Select the statistic of a variable to be displayed for a specified equipment, unit, and period.


[Select SPC rule] button Click this to select an SPC rule used for diagnosis.

Decision rule Shows the selected judgment type for the SPC rule.

SPC reset interval interval (sec) Set the interval between sections in seconds (0 to 10000) to reset the SPC rule count.

When '0' is specified, this setting will be disabled.

[Create SPC] button Click this to create the SPC specification value.

[Update Specs(GUI)] button Click this to edit SPC specification values.

[Update Specs(File)] button

[DIAG.] button Click this to start a diagnosis using the SPC rule.

(1)

(2)

(3)


21

*1 Only one of the following buttons can be selected: [Max], [Ave], [Min], [Range], [Center], [Sigma], [Integration], and [Length] buttons.





• [CV] button

• [Lock] button



• [3D] button

• Histogram

• [Var.] tab

• [EQ Def] tab

[Cpk] button Click this to display the Cpk chart when the SPC specification value is created by clicking the [Create SPC] button.

Page 204 Displaying Cpk chart

[Diagnosing] button This button is enabled when simple diagnosis (SPC diagnosis, multivariate analysis (single central value format), or

guard band diagnosis) is performed.

Click this button to open the "Simple diagnosis list under execution" screen.

Page 205 Displaying simple diagnoses being executed






When multiple variables are selected, they are displayed in order from the variable listed top in the "Select variable"

list.


The name of a legend can be changed arbitrarily, but it returns to the original name if selecting a variable again.

Item Description


Selecting SPC ruleSelect an SPC rule used for diagnosis. Default SPC rules are provided and they can be edited.


Page 193 Default SPC rule, Page 196 Editing SPC rule

Operating procedure

1. Click the [Select SPC rule] button to select an SPC rule.

The decision rule is shown in "Decision rule".


21

■ Default SPC ruleThe default SPC rules are shown below.

When the condition is satisfied, it is judged as an error.• 1 outside (2/5): outside of +- 1 sigma (2/5)

2 out of 5 successive points reside outside 1 .

• 2 outside (2/3): outside of +- 2 sigma (2/3)

2 out of 3 successive points reside outside 2 .


1 point resides outside 3 .





1σ

2σ

1σ

2σ

The condition (SPC rule) is satisfied.Upper management limit

Lower management limit

Center line

1σ

2σ

1σ

2σThe condition (SPC rule) is satisfied.

Upper management limit


Center line

3σ

3σ



Center line

4σ

4σ



Center line

5σ

5σ



Center line


• Management limit level 1(1/1): outside of control limits

Lv.1(1/1)

1 point resides outside the upper or lower management limit.

• Alternate increase and decrease (6/6): alternating

direction(6/6)

6 points alternately increase and decrease.

• Increase and decrease (6/6): increasing or decreasing(6/6)

Consecutive 6 points increase or decrease.

• Toward center (15/15): within one sigma(15/15)

15 successive points reside inside 1 .

• Continuous (9/9): plot on the same side(9/9)

Continuous 9 points reside on the same side of the center line.

1σ

2σ

1σ

2σ

The condition (SPC rule) is satisfied.



Center line

1σ

2σ

1σ

2σ




Center line

1σ

2σ

1σ

2σ



Center line


1σ

2σ

1σ

2σ




Center line

1σ

2σ

1σ

2σ




Center line


21

• Increase (10/10): increasing(10/10)

Increase (Continuous 10 points increase.)

• Decrease (10/10): decreasing(10/10)

Decrease (Continuous 10 points decrease.)

1σ

2σ

1σ

2σ



Center line

1σ

2σ

1σ

2σThe condition (SPC rule) is satisfied.



Center line


■ Editing SPC ruleThe contents of the SPC rules can be edited or multiple decision rules can be defined for one SPC rule by editing the contents

of 'SPCRule.csv'.

A new SPC rule can also be created.

Operating procedure

1. Click the [Update Specs] button.

The folder in which the SPC rules are stored is displayed.

2. Open 'SPCRule.csv' in the folder.

3. Change the value in "SPC rule name" column to an arbitrary name.

4. Change the value in the "Judgment type (n/m)1" column.

5. To specify multiple judgment rules, specify the judgment types in the "Judgment type (n/m)2" and later columns.

6. In the "EWMA& weighting factor" column, specify a weighting factor (smoothing factor) and the number of moving

averages of EWMA (Exponentially Weighted Moving Average).

7. When editing completes, save the file.

When clicking the [Select SPC rule] button, a name set for "SPC rule name" is displayed.

In addition, when selecting an SPC rule, a rule specified for "Judgment type (n/m) 1" to "Judgment type (n/m) 4" is displayed

in "Decision rule".


21

Precautions

• After editing the 'SPCRule.csv' file, Analyzer must be restarted.

• The numerical value of the judgment type 'n/m' can be changed to a value satisfying 'nm (n>0, m>0)'. For 'n' and 'm', set

the number of points (n) to define the timing of an error determination, and it is determined as an error when a condition is

satisfied by the specified number of points among consecutive points (m).

• Values of "n" and "m" can be specified within the range from 1 to 20. If a value out of this range is specified, the operation is

not guaranteed.

• Other parts of the judgment type than "(n/m)" cannot be changed from the following defaults:

outside of +- 1 sigma





outside of control limits Lv.1

alternating direction

increasing or decreasing

within one sigma

plot on the same side

increasing

decreasing

• If the judgment types (n/m)1 to (n/m)3 are blank, judgment types in the succeeding columns are ignored if any.

• For values to be specified in the "EWMA& weighting factor" column, integer values indicate the number of moving averages

and decimal values indicate EWMA weighting factors. When specifying both an integer value more than '1' and a decimal

value, the EWMA conversion will not be performed.

• When editing a CSV file in a text editor, be careful not to add or delete control symbols such as commas.

• The 'SPCRule_new.csv' file is stored in the same folder as the 'SPCRule.csv' file. This is the default file of 'SPCRule.csv' of

Real-time Statistic Diagnosis Tool which is currently being used. To initialize the 'SPCRule.csv' file after edited, delete the

'SPCRule.csv' file and rename the 'SPCRule_new.csv' file to 'SPCRule.csv'.

• To use a new standard SPC rule which is added when upgrading Real-time Statistic Diagnosis Tool, save the original

'SPCRule.csv' file and rename the copy of the 'SPCRule_new.csv' file to 'SPCRule.csv'. Then, apply the changes of the

saved 'SPCRule.csv' file as necessary. (The 'SPCRule.csv' file will not be changed with the update of Real-time Statistic

Diagnosis Tool.)

• When any of the following conditions are applied in the selected SPC rule, an error occurs. In this case, modify it in the

'SPCRule_new.csv' file

In the judgment type (n/m) 1 to 4, the value of 'n' is more than that of 'm' (n > m).

In the judgment type (n/m) 1 to 4, the value of 'n' or 'm' is less than '1'.

In the judgment type (n/m) 1 to 4, all of the judgement type are undefined or with incorrect names.

• Judgment is performed in order from the judgment type (n/m)1 to (n/m)4, and stopped when an error is

detected.

• There is no restriction on combinations of judgment types which can be defined in "Judgment type (n/m)1"

to "Judgment type (n/m)4".


Creating SPCCreate the SPC specification value.

SPC specification values can be modified by editing 'SPC specification value setting file'.


Page 199 Editing an SPC specification value

Operating procedure

1. Click the [Create SPC] button.

2. Statistic data of the period specified by "From" and "To" is used to calculate the standard deviation ( ), maximum value

(upper management limit), and minimum value (lower management limit), and SPC specification values ('SPC

specification value setting file') are created based on them.

3. When an error is detected, the mark is plotted for data with the error.

• Since the standard deviation is calculated from the read data, preparing data for sufficient sections ensures

creation of an accurate SPC standard value.

• SPC can also be created by specifying a filtering text name. In this case, a file different from the SPC

specification value setting file crated without specifying the filtering text name is created.

• Determine which judgment rule was used to detect the error from the color of plotted .

Judgment rule 1: Red

Judgment rule 2: Pink

Judgment rule 3: Yellow

Judgment rule 4: Light blue

Precautions

• The following two patterns of the SPC specification value setting file can be created (one for each) for a single variable:

When the filtering text name is specified

When the filtering text name is not specified

Therefore, if an SPC specification value setting file for the condition when the filtering text name is specified or is not

specified has already been created, the [Create SPC] button for the corresponding condition is disabled.

• When multiple variables are selected and a variable for which [Create SPC] has already been done for a certain condition

is included in them, the [Create SPC] button for that condition is disabled.

• To create SPC again, open a folder that stores an SPC specification value setting file by clicking the [Update Specs(File)]

button, delete the file, select a target variable again, then click the [Create SPC] button.


21

Editing an SPC specification valueThe SPC specification values can be modified by editing an SPC specification value setting file.

An SPC specification value setting file can be edited in the screens displayed by clicking the following buttons:

• [Update Specs(GUI)] button (Page 199 Updating specifications (GUI))

• [Update Specs(File)] button ( Updating specifications (file))

■ Updating specifications (GUI)The following shows the operating procedure for editing an SPC specification value setting file by clicking the [Update

Specs(GUI)] button.

Precautions

The [Update Specs(GUI)] button can be clicked when all the following conditions are satisfied:

• Only one variable is selected as a diagnosis target.

• An SPC rule is selected.

• An SPC specification value setting file for the diagnosis target variables is created.

• The "Filter display range by variable" screen*1 is not displayed.

*1 Appears by clicking the [FLT] button in the SPC screen.

Operating procedure

1. Click the [Update Specs(GUI)] button.

2. Edit SPC specification values (upper limit, target, lower limit, and sigma) in the "Update Specs(GUI)" screen, and click

the [OK] button.

The edited SPC specification values are applied to an SPC specification value setting file as well as to a graph in the SPC

screen.

Item Description

Upper limit Set the upper limit for the management value of statistics selected in the SPC screen.*1*2

By default, the value that is obtained by adding a margin*4 to the maximum value of the statistics for each section

included in the period specified for "From" and "To" of a selected variable is set.

Target Set the value to be used as the center line of statistics selected in the SPC screen.*1*2

By default, the average value of the statistics for each section included in the period specified for "From" and "To" of a

selected variable is set.

Lower limit Set the lower limit for the management value of statistics selected in the SPC screen.*1*2

By default, the value that is obtained by subtracting a margin*4 from the minimum value of the statistics for each section



*1 Available number of characters: 1 to 22 characters (significant digits of mantissa: 15 digits)*2 Available numeric values: 0, 2.22507385850721E-308 (absolute value) to 1.79769313486231E308 (absolute value).*3 Available numeric values: 0, 2.22507385850721E-308 to 1.79769313486231E308*4 The margin is 1% of the width from the minimum value to the maximum value of statistic.

Sigma Set the standard deviation of statistics selected in the SPC screen.*1*3

By default, the standard deviation (1 sigma) of the statistics for each section included in the period specified for "From"

and "To" of a selected variable is set.

[Default] button Click this to restore the default values for the SPC specification values in the "Update Specs(GUI)" screen.

Item Description


21

■ Updating specifications (file)The following shows the operating procedure for editing an SPC specification value setting file by clicking the [Update

Specs(File)] button.

Operating procedure

1. Click the [Update Specs(File)] button.

2. Select an SPC specification value setting file.

An SPC specification value setting file name is assigned as follows:

File name*1: UNIT_(unit name)_(filtering text name)_SPC creation target variable name.csv

*1 When a filtering text name is not specified, it is not displayed.

3. Edit the specification values of "Upper limit," "Lower limit," "Target," and "Sigma" for each statistics data item.

*1 The margin is 1% of the width from the minimum value to the maximum value of statistic.

Precautions

• When editing a CSV file in a text editor, be careful not to add or delete control symbols such as commas.

Item Description

Upper limit Set the upper limit for the management value.

By default, the value that is obtained by adding a margin*1 to the maximum value of the statistics for each section


Lower limit Set the lower limit for the management value.

By default, the value that is obtained by subtracting a margin*1 from the minimum value of the statistics for each section


Target Set the value to be used as the center line.

By default, the average value of the statistics for each section included in the period specified for "From" and "To" of a

selected variable is set.

Sigma Set the standard deviation.

By default, the standard deviation (1 sigma) of the statistics for each section included in the period specified for "From"

and "To" of a selected variable is set.


Starting diagnosis using SPC ruleDiagnosis starts using the specified SPC rule.

Operating procedure

1. Click the [DIAG.] button.

2. Click the [Diagnostics start] button.

Clicking the [DIAG.] button [Read set value] button reads a setting being diagnosed or stopped.

Precautions

The settings for Monitor must be set in advance to cooperate with Real-time Flow Manager when an error is detected.

Page 416 Monitor


21

Updating diagnosis definitionIf the diagnosis rule is changed after starting diagnosis, the changed rule is used.

Operating procedure


2. Click the [Diagnostics definition update] button.

Stopping diagnosisStop diagnosis.

Operating procedure


2. Click the [Stop diagnostics] button.


Displaying Cpk chartA Cpk chart is displayed in the trend chart (Statistics) display screen when the SPC specification value is created by clicking

the [Create SPC] button.

Cpk (process capability index) represents how accurately the production line (process) can manufacture products by an index

(numeric value).

The Cpk value allows for evaluating the process capability.


Page 204 Evaluating process capability by Cpk value

Operating procedure

1. Click the [Cpk] button.

2. Set each item.

3. A Cpk chart is shown.

mark is plotted for data lower than the specified lower limit of Cpk.

Cpk can be calculated by the following equation:

• Cpk = min{((Upper management limit) - (Average value of specified section)) / 3( of specified section),

((Average value of specified section) - (Lower management limit)) / 3( of specified section)}

■ Evaluating process capability by Cpk valueThe Cpk value allows for evaluating the process capability as follows:

Item Description

Num of Calc-sections. Enter the number of sections used for calculating Cpk (2 to 1000).

Lower limit Enter the lower limit value of Cpk (0.0000 to 10000.0000).

Cpk value Description

Cpk1.67 Process capability is more than sufficient and there is almost no possibility to produce non-conforming products.

A little variation of characteristic values including long-term fluctuation will not be a problem.

It the current management is too strict, it can be mitigated.

1.67>Cpk1.33 Process capability is sufficient and a problem will not occur if the process is stabilized within this performance range

and proper maintenance and management are provided.

1.33>Cpk1.00 Process capability exists but is not sufficient.

For a critical characteristic, improvement is required to raise Cpk to 1.33 or higher.

Continuous improvement is necessary to ensure Cpk is maintained at 1.33 or higher even for the internal

specifications.

1.00>Cpk Process capability is insufficient and it is necessary to inspect all produced products. This results in increase of costs

due to additional inspection and yield loss caused by non-conforming products, and requires quick improvement to

raise Cpk to 1.00 or higher.


21

Displaying simple diagnoses being executedBy clicking the [Diagnosing] button during simple diagnosis, diagnoses being executed are displayed in a list. In addition,

diagnoses being executed can also be stopped in the list.

Window

Displayed items

Item Description

Select The selected status of the execution target for the [Stop diagnostics] button is displayed.

• Selected: Execution target

• Not selected: Not execution target

Diagnosis name The names of simple diagnoses being executed are displayed.

Diagnosis start date The start date (YYYY/MM/DD hh:mm:ss) of simple diagnosis being executed is displayed.

Type The type of simple diagnosis is displayed.

• SPC: SPC

• Multivariate analysis (single central value format) (T2): Multivariate analysis (single central value format) (When "T2"

is selected for "Calc type")

• Multivariate analysis (single central value format) (Q): Multivariate analysis (single central value format) (When "Q" is

selected for "Calc type")

• Multivariate analysis (single central value format) (KNN): Multivariate analysis (single central value format) (When

"KNN" is selected for "Calc type")

• Multivariate analysis (single central value format) (MT): Multivariate analysis (single central value format) (When

"MT" is selected for "Calc type")

• Multivariate analysis (single central value format) (EU): Multivariate analysis (single central value format) (When

"EU" is selected for "Calc type")

• Guard band: Guard band

[Select all] button This button is enabled when there is an item in the list or an item whose checkbox in the "Selection" column is not

selected.

Click this to select all checkboxes in the "Selection" column of simple diagnoses.

[Unselect all] button This button is enabled when there is an item whose checkbox in the "Selection" column is selected.

Click this to unselect all checkboxes in the "Selection" column of simple diagnoses.

Display type The type of simple diagnosis to display in the list can be selected.

It is selected by default according to the simple diagnosis type of the screen where the [Diagnosing] button is clicked.

• All: All types of simple diagnoses are displayed.

• SPC: The simple diagnoses using SPC are displayed.

• Multivariate analysis (single central value format): The simple diagnoses using Multivariate analysis (single central

value format) are displayed.

• Guard band: The simple diagnoses using Guard band are displayed.

[Diagnosis stop] button This button is enabled when there is an item whose checkbox in the "Selection" column is selected.

Click this to stop diagnosis for the item whose checkbox in the "Selection" column is selected.

If multiple checkboxes in the "Selection" column are selected, they are stopped one by one.


■ When simple diagnosis cannot be stoppedThe following screen appears when simple diagnosis being executed cannot be stopped.

Window

Displayed items

Item Description

[Abort] button Click this to cancel stopping simple diagnoses succeeding to the diagnosis which could not be stopped.

[Retry] button Click this button to retry stopping the simple diagnosis which could not be stopped.

[Ignore] button Click this button to skip the simple diagnosis which could not be stopped.

The next simple diagnosis will be stopped if there is another diagnosis to be stopped.

21 Analyzer21.5 Multivariate Analysis (Single Central Value Format) 207

21

21.5 Multivariate Analysis (Single Central Value Format)

Statistic calculation of each variable for each section of CSV file data read by Reader, multivariate analysis of statistics,

creation of a reference waveform, and diagnosis based on the created reference waveform can be performed with simple

operations.

Operating procedure

1. Select [Simple analysis - Simple diagnosis] [MVA] [Single central value format].



3. Create a reference waveform for diagnosis.

Page 211 Creating the reference waveform

4. Start diagnosis.

Page 214 Starting diagnosis



6. Stops diagnosis.


20821 Analyzer21.5 Multivariate Analysis (Single Central Value Format)

Screen configurationThe following shows the screen configuration for multivariate analysis (single central value format).

Displayed items

Item Description

(1) Trend chart (statistics) display screen Shows the statistics for a variable selected from the "Select variable" list.

• The vertical axis is automatically scaled according to the maximum and minimum values of the displayed

waveform. Note that when multiple variables are selected, the vertical axis values are not displayed.

• Numbers corresponding to the number of units of data (sections) included in the period specified for "From" and

"To" are displayed on the horizontal axis.

(2) Histogram/scatter chart/trend graph

(reference waveform) display screen

Shows the histogram for a variable selected from the "Select variable" list.

When two or more variables are selected from the "Select variable" list, the scatter chart is shown.

By clicking the [Make criterion] button, a waveform to be referenced of diagnosis is displayed.

Page 211 Creating the reference waveform

(3) Calculation result display screen Shows the result of calculation performed on the reference waveform according to the selected calculation type.

Select variable Select the statistic of a variable to be displayed for a specified equipment, unit, and period.




When selecting the section data in the section list while the cursor is displayed, the waveform of the section is

snapped by the cursor. In addition, when snapping a waveform by the cursor, the corresponding section data in

the section list is selected.

When hiding the data of all sections in the section list, the statistics of all sections before narrowing down or hiding

is displayed.








[Range] button*1 Click this to display the difference between the maximum and minimum values of a waveform for each section.




(1)

(2)

(3)


21

*1 Only one of the following buttons can be selected: [Max], [Ave], [Min], [Range], [Center], [Sigma], and [Integration] buttons.*2 It is displayed when "T2" or "Q" is selected for "Calc type".*3 Shown when "KNN" is selected for "Calc type".





• [Multi] button

• [CV] button

• [Lock] button



• [3D] button

Calc type Select the calculation type.

T2 Diagnosis is performed using the Hotelling T2 (Mahalanobis) distance as the upper limit value.

Multicollinearity errors can be avoided.

The number of principal components needs to be specified for calculation.

It is assumed that the Gauss distribution is followed.

Q Diagnosis is performed using the Q statistic as the upper limit value.

Multicollinearity errors can be avoided.

The number of principal components needs to be specified for calculation.


KNN Diagnosis is performed using the standardized Euclid distance as the upper limit value.

Calculates the average distance from the points in the reference for the number of Ks closest to the diagnostic

point.

Applied when the distribution includes a non-linear pattern.

MT Diagnosis is performed using the MT (Mahalanobis-Taguchi) distance as the upper limit value.

A multicollinearity error may occur between highly correlated variables


EU Diagnosis is performed using the standardized Euclid distance from the center of gravity of the criterion space

(criterion matrix) as the upper limit value.

Standardized Euclid distance = ((diagnosis value - average of criterion matrix) / of criterion matrix)2

It is applied when variables are totally not correlated.

Upper limit Set the upper limit value (0.0000 to 1000.0000) for diagnosis.

PC. num*2 Set the number of principal components (1 or 2).

K number*3 Set the number of Ks (1 to 1000).

Fault judgement sections Set the number of times of upper limit excess. When the set number is exceeded, an error is detected.

Enter the number of error occurrences (1 to 1000) in the left text box and the number of determination sections (1

to 1000) in the right text box.

(Example) When entering '2' in the left text box and '5' in the right text box:

It is regarded as an error when the upper limit value is exceeded for two or more times in consecutive five

sections.

[Make criterion] button Click this to create the reference waveform.

[Read criterion] button Click this to read the saved reference waveform.

For the path of a file to be read, up to 255 characters (two characters are used for a double-byte character)

including the extension can be specified.

[Save criterion.] button Click this to specify a reference name (up to 255 characters) and save the reference waveform.

Setting parameters are not saved.

When saving a reference waveform, the following prefixes are set for the file name corresponding to the settings

when creating the reference waveform. Set a name after the prefixes since the prefixes cannot be changed. (If

they have been changed, an error will occur and the reference waveform cannot be saved.)

• UNIT_(unit name)_(statistics name)_(filtering text name)_(filtering text value)_

If the file is saved in a folder other than 'Reader equipment data folder' and 'RSD dataset folder, the file needs to

be moved manually to the migration destination when migrating the settings to Real-time Statistic Diagnosis Tool

installed in another industrial PC.

Page 121 Migrating settings to Real-time Statistic Diagnosis Tool in another industrial PC, Page 122 Project

File Setting Utility

[DIAG.] button Click this to start diagnosis in multivariate analysis (single central value format).

[Diagnosing] button This button is enabled when simple diagnosis (SPC diagnosis, multivariate analysis (single central value format),

or guard band diagnosis) is performed.



Item Description


• Histogram

• [Var.] tab

• [EQ Def] tab


21

Creating the reference waveformCreate a reference waveform for diagnosis.

Operating procedure

1. Select multiple variables in the "Select variable" list. Two or more variables must be selected to create the reference

waveform.

2. The scatter diagram is shown on the histogram/scatter chart/trend graph (reference waveform) display screen.

Probability ellipses for 99%, 95%, and 90% and regression lines are displayed. When correlation exists with the hazard rate of

5%, the background is shown in blue.

3. Click the [Make criterion] button. The reference waveform is displayed in the histogram/scatter chart/trend graph

(reference waveform) display screen.

The number of sections of the reference waveform must be equal to or more than the number of the selected

variables.


4. Set the calculation type and parameters.

For details on the calculation types and parameters, refer to the following:


5. The calculation result is displayed in the calculation result display screen.

6. When "Upper limit" is less than the maximum value, a marker (reference waveform: yellow, calculation result: red) is

displayed at each portion exceeding the maximum value on the histogram/scatter chart/trend graph (reference

waveform) display screen and the calculation result display screen.

• The settings in "Fault judgement sections" are not applied to the error markers displayed in the histogram/

scatter chart/trend graph (reference waveform) screen. (All error markers are displayed.)

• For the error markers displayed in the 'Calculation result display' screen, the settings in 'Fault judgement

sections' are applied.


21

7. Right-click on the calculation result display screen and select [+], [|], or [--] from [Cursor] to display the cursor.

By snapping a graph using a cursor, the variable deviation graphs for each variable in the snapped section are displayed.

Variable deviation can be calculated by the following equation.

Variable deviation = (variable value in a snapped section - average value of the reference waveform of the

variable) / (standard deviation of the reference waveform of the variable)

8. Right-click on the histogram/scatter chart/trend graph (reference waveform) display screen and select [Delete error

point].

Delete error sections from the histogram/scatter chart/trend chart (reference waveform) display screen.

When the error sections are deleted, the calculation result display screen is also updated. Determine the

upper limit and start diagnosis again with the markers hidden.

Precautions

• Only one reference waveform is created for each unit. In addition, a reference waveform is created for a selected variable

and a selected section.

• When a variable of which the statistic does not change is selected, the reference waveform for the variable will not be

created.

• When performing a diagnosis after deleting an abnormal section from the reference waveform, click the [DIAG.] button

without clicking the [Make criterion] button. If the [Make criterion] button is clicked, the deleted abnormal section will be

restored.


Starting diagnosisStart diagnosis according to the set parameters.

Operating procedure



Clicking the [DIAG.] button [Read set value] button reads a setting being diagnosed or stopped.

Precautions

• When starting diagnosing, saving or reading the reference waveform (by clicking [Save criterion.] or [Read criterion] button)

is required.

• The settings for Monitor must be set in advance to cooperate with Real-time Flow Manager when an error is detected.

Page 416 Monitor

21 Analyzer21.6 Guard Band Diagnosis 215

21

21.6 Guard Band DiagnosisCreate a guard band to perform a diagnosis by determining the range of values regarded to be normal based on the CSV file

data read by Reader.

The following two cases are possible for performing a diagnosis using a guard band.

• Diagnosis by creating a guard band based on the reference waveform

• Performing diagnosis by setting the upper and lower limits

For a reference waveform, multiple sections can be selected.


(2) Create a guard band

(3) No error

(4) Error

(a) Guard band upper limit

(b) Guard band lower limit

(c) Error domain


(2) Set the upper and lower limits

(3) No error

(4) Error

(a) Upper Limit

(b) Lower Limit

(c) Error domain

(1)(2)

(3)

(4)

(a)

(b)

(c)

(1) (2)

(3)

(4)

(a)

(b)

(c)

21621 Analyzer21.6 Guard Band Diagnosis

Operating procedure

1. Select [Simple analysis - Simple diagnosis] [Guard band].



Only one variable can be selected.

3. Select the guard band type.

Page 220 Selecting guard band type

4. Create a guard band.

Page 220 Creating guard band

5. Start diagnosis.

Page 221 Starting guard band diagnosis



7. Stops diagnosis.



21

Screen configurationThe following shows the screen configuration for a guard band diagnosis.

Displayed items

Item Description

(1) Trend graph (waveform) display screen Shows the waveform for a variable selected from the "Select variable" list.

• The vertical axis is automatically scaled according to the maximum and minimum

values of the displayed waveform.

• The horizontal axis shows the number of seconds from the start point of the

section.

The display status of the waveform currently snapped (grabbed) by the cursor can

be switched to hide.


(2) GB evaluation result display screen Shows the result of diagnosis performed by clicking the [Eval-GB.] button for data

during the period specified by the period specified for "From" and "To".

Select variable Select a variable to be displayed for a specified equipment, unit, and period.

Multiple variables cannot be selected.

MA (noise removal) Set when calculating the moving average of the waveform, rejecting noises, and

creating a GB waveform.

Specify the number of points to calculate the moving average.

[GB] tab Set parameters used for creating a guard band based on a waveform.

Specify a numeric value or percentage for each parameter to create a guard band.

V width Specify a value (0.0000 to 2147483647.0000) that indicates how much a new guard

band shifts up or down along the Y axis of a reference waveform.

This setting is effective when the "V width (%)" value is '0'.

(3) V width (%) Specify the percent of the maximum amplitude (0.000 to 100.0000) that indicates

how much a new guard band shifts up or down along the Y axis of a reference

waveform.

This setting is effective when its value is not "0".

F/R width Specify a value (0 to 2147483647) that indicates how much a new guard band shifts

left or right along X axis of a reference waveform.

This setting is effective when the "F/R width (%)" value is '0'.

(4) F/R width (%) Specify the maximum section length (0.000 to 100.0000) that indicates how much a

new guard band shifts left or right along X axis of a reference waveform.

This setting is effective when its value is not "0".

Median Creates a guard band using the median of the displayed waveform as the reference

waveform.

(2)

(1) (5)

(6)

(3)(4)


[Upper and lower limit] tab Set parameters used for creating a guard band by specifying the upper and lower

limits.

Enter a number so that the upper limit value is bigger than the lower limit value.

Upper limit Specify the upper limit value (-2147483648.0000 to 2147483647.0000).

Lower limit Specify the lower limit value (-2147483648.0000 to 2147483647.0000).

Continuous excess point Specify the number (1 to 2147483647) of consecutive excess points to determine

that a value deviates from the guard band. A point in which a value matches with

the guard band is also regarded it deviates the guard band.

Start offset Starting from the start point of a section, specify the positive number of points (0 to

2147483647) at which the guard band is started.

End offset Specify a number (-2147483648 to 0) of point at which the guard band is ended

starting from the section end point.

Fault judgement sections Specify the total number of consecutive sections and what number section at which

a deviation occurs out of the specified consecutive sections to define a diagnosis

result as an error. It is regarded as an error when the guard band deviates in the

specified section out of the specified number of consecutive sections.

Set the numerator (1 to 1000) in the left text box and the denominator (1 to 1000) in

the right text box. (Integer only)

(Example) When '2' is entered in the left field and '3' is entered in the right field

It is judged to be an error when two sections out of three consecutive sections

deviate from GB.

[Make GB] button Create a guard band.

[Eval-GB.] button Diagnoses data during the period specified by the period specified for "From" and

"To" and shows the result in the GB evaluation result area.

[Read GB] button Reads the saved guard band setting value (save file*1) by selecting the value.

For the path of a file to be read, up to 255 characters (two characters are used for a

double-byte character) including the extension can be specified.

[Save GB] button Saves the setting value of the guard band in the format of Real-time Data Analyzer

or programmable controller.

Page 221 Saving the guard band

• The default save file name is shown below. (It can be changed to any name within

255 characters including an extension.)

• UNIT_(unit name)_(variable name selected in "Select variable")_(specified

filtering text name)_(text value selected for filtering text specification).GB

• For the path of a file to be saved, up to 255 characters (two characters are used

for a double-byte character) including the extension can be specified.

• If the file is saved in a folder other than 'Reader equipment data folder' and 'RSD

dataset folder, the file needs to be moved manually to the migration destination

when migrating the settings to Real-time Statistic Diagnosis Tool installed in

another industrial PC.

Page 121 Migrating settings to Real-time Statistic Diagnosis Tool in another

industrial PC, Page 122 Project File Setting Utility

[DIAG.] button Diagnosis starts using the guard band.

(5) Cursor legend The legend of the waveform currently snapped (grabbed) by the cursor can be

displayed.




(6) Chart legend Shows the legend of the waveform, upper limit, or lower limit chart.


The name of a legend can be changed arbitrarily, but it returns to the original name

if selecting a variable again.

[Diagnosing] button This button is enabled when simple diagnosis (SPC diagnosis, multivariate analysis

(single central value format), or guard band diagnosis) is performed.



Item Description


21

*1 Only files in the Real-time Data Analyzer format can be read.




• [Multi] button


• [3D] button

• [CV] button

• [FLT] button

• [Lock] button


• [Max] button

• [Min] button

• [Ave] button

• [Range] button

• [Center] button

• [Sigma] button



• Histogram

• [Var.] tab

• [EQ Def] tab


Selecting guard band typeSelect the guard band type.

Operating procedure

1. Click the [GB] or [Upper and lower limit] tab depending on the type of the guard band to create.

2. Set the parameters for creating the guard band.

For the parameters, refer to the following:


Creating guard bandCreate a guard band.

Operating procedure

1. Click the [Make GB] button.

The guard band is created.

When snapping a waveform (reference waveform or guard band) by the cursor, the data number of the

snapped area (the number counted from the first waveform) and the number of data points of the entire

waveform will be displayed on the upper left in the trend graph (waveform) display screen as follows.

• (data number) / (number of data points)


21

Starting guard band diagnosisDiagnosis starts using the created guard band.

Operating procedure



Clicking the [DIAG.] button [Read set value] button reads a setting being diagnosed.

Precautions

• If the [Diagnosing] button is clicked and the diagnosis file is deleted, the set values are also deleted and cannot read even

by clicking the [Read set value] button.

• The settings for Monitor must be set in advance to cooperate with Real-time Flow Manager when an error is detected.

Page 416 Monitor

Saving the guard bandSave upper and lower limit waveforms of a created guard band.

Operating procedure

1. Click the [Save GB] button.

The created guard band is saved.

■ Save formatThe following table shows the format when saving a guard band.

When the extension not corresponding to the selected format is specified, the correct extension is

automatically added and the guard band is saved.

Format (extension) Description

Real-time Data Analyzer format (.GB) This format is available in Real-time Data Analyzer.

Saved guard bands can be read by using the [Read GB] button.

Programmable controller format (.PGB) This format is available in a programmable controller.

For details on this format, refer to the following:

Page 394 Guard band diagnosis

22221 Analyzer21.7 Creating Correlation Matrix

21.7 Creating Correlation MatrixCorrelation analysis can be performed by reading a CSV file offline and creating a correlation matrix.

The following lists types of CSV files which can be used to create a correlation matrix.

A created correlation matrix can be used in the following sections.



The CSV file must contain the column item title on the first line and data on the second and later lines. (No

time stamp column is required.)

Operating procedure

1. Select [Advanced analysis] [Create correlation matrix].

2. Create a whole matrix.

Page 225 Creating whole matrix

3. Delete incorrect variables from the created whole matrix.

Page 227 Deleting an incorrect variable

4. Add measurement data to the whole matrix.

Page 228 Adding measurement data to a whole matrix

5. Create a partial matrix.

Page 231 Creating partial matrix

6. Display and analyze the correlation graph.

Page 235 Correlation analysis by using a correlation graph

Classification Description

Create in Real-time Statistic

Diagnosis Tool

CSV file created by clicking the [CSV for matrix] button on the following function screens:




Page 189 SPC


CSV file created by extracting statistics and saving them in the CSV format in 'Editing Logics'


CSV file created by saving variable values in the CSV format in 'Editing Logics'


CSV files created manually CSV file created according to the standard format

21 Analyzer21.7 Creating Correlation Matrix 223

21

Screen configurationThe following shows the screen configuration for creating a correlation matrix.

Item Description

[Whole matrix] tab Read a CSV file and create and operate the whole matrix.

Matrix storage folder*1 Specify a folder path to save a matrix (up to 256 characters).

If the file is saved in a folder other than 'Reader equipment data folder' and 'RSD

dataset folder, the file needs to be moved manually to the migration destination

when migrating the settings to Real-time Statistic Diagnosis Tool installed in another

industrial PC.

Page 121 Migrating settings to Real-time Statistic Diagnosis Tool in another

industrial PC, Page 122 Project File Setting Utility

[Subfolder] button When creating a subfolder for each equipment in a matrix storage folder specified

for "Matrix storage folder", select the subfolder.*1

Reading folder*1 Specify the folder path in which a CSV file containing equipment data is stored (up

to 256 characters).

Up to ten folders can be registered or selected.

CSV file name Shows the name of the CSV file saved in the folder specified in "Reading folder".

[Device data reading] button Reads the highlighted CSV file(s) as the whole matrix.

[Read measurement data.] button Inserts measurement data into the created whole matrix.

Please input Matrix name first. Enter the name of a whole matrix to be displayed in "Whole matrix name" (up to 32

characters).

Whole matrix name Shows the existing whole matrices.

Highlight a whole matrix name to display the tab for that whole matrix.

[Data] tab Shows the data value of the matrix variable selected "Mat. variable name list".

[Histogram] tab Shows the histogram of the matrix variable selected "Mat. variable name list".

[NormalPrabability] tab Shows the normal probability chart of the matrix variable selected "Mat. variable

name list".

The vertical axis of the normal distribution chart shows the predicted accumulated

probability (%) while the horizontal axis shows the observed accumulated

probability (%).

[Trend] tab Shows the trend chart of the matrix variable selected "Mat. variable name list".

[Comment] tab Shows a comment automatically added when reading a file by clicking the [Device

data reading] button.

[Whole matrix] tab Shows the created whole matrices.


*1 Do not specify a folder with the encryption attribute enabled.

[Select and delete incorrect

variable] button

Detects and deletes incorrect variables in the matrix.

Delete highlight variable. Deletes highlighted (selected) variables and creates a matrix.

Columns of the matrix contain variable names and rows contain points (samples).

Highlight the variable which the

value doesn't change.

Highlights (selects) variables which have unchanging values.

Variables with a constant value or no value are excluded from a correlation analysis.

A constant value can be read if the variable is read as a string.

Highlight the variable that the

number of points doesn't come up

to the number of variables.

Highlights (selects) variables which have the less number of points than the number

of variables.

When the number of points (sample number) is less than the number of variables, a

correlation analysis can be performed, but Multiple regression analysis and

Maharanobis-Taguchi method can not be performed.

Retrieve variable in which matrix is

empty

Searches for and highlights (selects) a variable which has data only in one of the

same rows of two or more variables (i.e., an exclusive variable).

An exclusive variable causes the matrix to become empty and a partial matrix

cannot be created.

Retrieve variable of minimum

points

Highlights (selects) a variable which has the least number of points (sampling

count).

When any of variables has extremely few number of points, the number of points of

the matrix is adjusted to that (extremely few) number of points.

Mat. variable name list Shows the matrix variable names.

Item Description


21

Creating whole matrixA whole matrix to be a base of correlation analysis can be created by reading equipment data (data to be subjected to

correlation analysis) in the CSV file format.

Operating procedure

1. Specify a folder path to save a created whole matrix in "Matrix storage folder" and a folder path in which a CSV file used

for creating a correlation matrix is stored in "Reading folder" (up to 256 characters).

Up to ten folders can be specified in the "Reading folder".

2. When a folder is specified in the "Reading folder", the names of the CSV files saved in the specified folder is listed in

"CSV file name".

3. Select a CSV file(s) to create the whole matrix and click the [Device data reading] button.

4. Enter the whole matrix name (up to 32 characters) and click the [OK] button.

5. Select a numeric value(s) or string variable(s) to filter the matrix from "String column and filtering item".

Select a corresponding variable in "Numerical value item" for a variable used as a numerical value among variables used for

narrowing down.

To merge two or more variables into a single variable, select the merge target variables from "Merge item".

6. After selecting necessary items, click the [OK] button.


7. The whole matrix is created.

The whole matrix is saved in the folder specified in "Matrix storage folder" with the name '(input whole matrix name).csv'.

When a subfolder is specified, the whole matrix is saved in the subfolder selected by clicking the [Subfolder] button.

8. Apply Gauss noise to data of the created whole matrix to give variation.

Right-click the matrix name in "Whole matrix name" and select [Add Gauss noise].

9. Select the name of a variable to be added a Gauss noise and enter sigma (0.0000 to 100.0000) of a noise in "Gauss

noise (sigma)" to add the noise.

10.After adding the Gauss noise, select all variables, enter a name (up to 32 characters) different from the original name of

the whole matrix in "Variable matrix name", and click the [Save] button.

The whole matrix with the Gauss noise added is created.


21

Precautions

• Do not specify a folder with the encryption attribute enabled at the "Matrix storage folder", "Reading folder", and the

subfolder selected by clicking the [Subfolder] button.

• When the size of a specified CSV file is too large, a memory overflow error occurs and the file may not be read. In this case,

reduce the file size by reducing the number of variables or that of records in the file. (It is recommended to reduce the file

size to 10 M bytes or less.)

• If the specified CSV file is too large, a memory overflow error occurs and the matrix may not be read or deleted. In this

case, delete the following file or folder:

Folder specified in "Matrix storage folder"*1\(matrix name).csv

Folder specified in "Matrix storage folder"*1\Attribute folder/(matrix name folder)

*1 When a subfolder is specified, the subfolder selected by clicking the [Subfolder] button is used.

• If multiple CSV files are selected to create the whole matrix, they are not read unless all of the column item names (on the

first row) are identical in all files.

• When setting a folder path in "Matrix storage folder," set the path so that the path length of a setting file to be created in the

set folder does not exceed 259 characters. The maximum path length of a setting file is as follows:

(Set folder)\ + 113 characters (if the whole matrix name and partial matrix name are 32 characters long each)

• When creating a whole matrix with the Gauss noise added, data for the number of sections more than the number of

selected variables are required.

• When creating a whole matrix with the Gauss noise added, it cannot be saved as the same name of the original whole

matrix (i.e., overwrite it).

• Variables not selected in "String column and filtering item" are regarded as numeric values.

• The variable values are merged as the value of the variable listed top. Same data is copied except for variables targeted for

merge.

Deleting an incorrect variableDelete an incorrect variable such as ones which are not required for a correlation analysis from a created whole matrix.

Operating procedure

1. Click the [Select and delete incorrect variable] button, and select an operation content.

For operations, refer to the following:


(1) Before merge

(2) After merge

(a) ID1, ID2, and ID3 are selected as merge items.

(a)

Value1 Value2 ID1 ID2 ID3

10 20

20

30

40

60

1

4

7

2

5

8

3

6

9

(1)

Value1 Value2 ID1

10 20 1

10 20 2

10 20 3

20 40 4

20 40 5

20 40 6

30 60 7

30 60 8

30 60 9

(2)


Adding measurement data to a whole matrixWhen CSV format measurement data such as quality evaluation data obtained by a product inspection after processing

products exists, the data can be added to a created whole matrix.

By linking measurement data and equipment data, more detailed correlation analysis can be performed.

This operation is necessary when the measurement data has been saved in a different file from equipment data.

Operating procedure

1. Select the name of a whole matrix, to which measurement data will be added, in a list, then specify the folder, in which

the measurement data has been stored, for "Reading folder".

If the measurement data and equipment data are stored in different folders, select and specify a tab different from that for the

equipment data in "Reading folder".

2. Select a CSV file of measurement data to be added (multiple selection available), and click the [Read measurement

data.] button.


4. Match the equipment data and measurement data.

Select a string item(s) of the measurement data in "Measurement data item". Select the equipment data and measurement

data items (up to three items) used for matching, and click the [OK] button.

For details of matching between the measurement data and equipment data, refer to the following:

Page 229 Matching between equipment data and measurement data


21

5. The measurement data is added in "Mat. variable name list".

A measurement data name is as follows. Equipment data and measurement data can be distinguished in

"Mat. variable name list" based on a measurement data name. In addition, it can be identified which tab the

measurement data is read from based on an added tab number.

• (Item name) + (read folder name) + (m) + (tab number in "Reading folder" (1, 2, 3))

Precautions

• Do not specify a folder with the encryption attribute enabled as the "Reading folder"

• When the size of a specified CSV file is too large, a memory overflow error occurs and the file may not be read. In this case,

reduce the file size by reducing the number of variables or that of records in the file.

(It is recommended to reduce the total size of the files specified for equipment data or measurement data to 10 M bytes or

less.)

■ Matching between equipment data and measurement dataMatching is performed using the matching keys from the bottom row toward the top row of the equipment data whole matrix

and measurement data matrix.

If the same matching key appears more than once, the equipment data and measurement data matching for the first time are

only paired.

The following shows the matching process with the equipment data and measurement data listed below.

• Equipment data

• Measurement data

Operating procedure

1. Select "RecID" and "LotID" as the key items of the measurement data.

2. Select 'ID1' and 'ID2' as the key items of the equipment data.

3. Match "RecID" with "ID1" and "LotID" with "ID2".

4. The fifth row of the measurement data matches the fifth row of the equipment data, and they are merged.

(1) Equipment data

(2) Measurement data

ID1Value1 Value2 ID2ZZZ111AAA11110 20ZZZ222AAA22220 40ZZZ333AAA33330 60ZZZ444AAA44440 80ZZZ555AAA55550 100

Value3 Value4 RecID LotIDZZZ111AAA11112 21ZZZ222AAA22324 42ZZZ333AAA33331 65ZZZ555AAA55541 81ZZZ555AAA55550 106


(1)Value3 Value4 RecID LotID

ZZZ111AAA11112 21ZZZ222AAA22324 42ZZZ333AAA33331 65ZZZ555AAA55541 81ZZZ555AAA55550 106

(2)


5. The fourth row of the measurement data matches the fifth row of the equipment data. Since the fifth row of the device

data has already been matched with the fifth row of the measurement data, the fourth row of the measurement data is

discarded.

6. The third row of the measurement data matches the third row of the equipment data, and they are merged.

7. Since 'RecID' on the second row of the measurement does not match anything under 'ID1' of the equipment data, the

second row of the measurement data is discarded.

8. The first row of the measurement data matches the first row of the equipment data, and they are merged.

9. Matching completes.

• Up to three items can be selected for matching.

• If the same matching key appears more than once, the logic uses the newer key. To use this function, the

equipment data and measurement data files must be arranged in the ascending order of the file names as

well as the rows in each file must be arranged in the order of timestamps. CSV files created by 'Logic

editing' automatically satisfy this requirement.


(1) Equipment data


(1) Equipment data


(1) Equipment data


(1) Equipment data





(2)




(2)




(2)




(2)

Value1 Value2211210 20

20 40653130 60

40 801065050 100

ID1 ID2ZZZ111AAA111ZZZ222AAA222ZZZ333AAA333ZZZ444AAA444ZZZ555AAA555

Value3 Value4


21

Creating partial matrixWhen a whole matrix is created, a partial matrix with the same name and contents as the whole matrix is created by default.

The created partial matrix is saved under the following folder path:

• (Folder specified for "Matrix storage folder")*1\Attribute\(whole matrix name folder)\(whole matrix name).csv


Operating procedure

1. Click a tab of the created partial matrix.

2. Edit a partial matrix using one of the following three methods:

• Filtering variables by text variable values ( Page 232 Creating partial matrix by filtering by text variable values)

• Deleting outliers or values other than outliers in the histogram ( Creating partial matrix by deleting outliers or

values other than outliers in the histogram)

• Deleting points in the trend chart ( Page 234 Creating partial matrix by deleting points in the trend chart)

Precautions

The partial matrix in the "Partial matrix name" list with the same name as the whole matrix is copied from the whole matrix and

the partial matrix is created based on it. This copy of the whole matrix cannot be changed or overwritten.

Item Description

Partial matrix name (It is possible

to save it also with Enter key).

Set a name to create a partial matrix (up to 32 characters).

Partial matrix that belongs to

selected whole matrix.

Displays the created partial matrix.

[Filtering state variable name]

button

Specify a text variable when filtering data with a text variable value.

(1) Section list of partial matrices The section list of partial matrices is displayed.

(1)


■ Creating partial matrix by filtering by text variable values

Operating procedure

1. Click the [Filtering state variable name] button to specify a text variable used to filter data.

Text variable values are displayed in a list.

2. Select a text variable value in the text variable value list.

Rows corresponding to a selected text variable value are highlighted in the section list of partial matrices and the [Data] tab.

3. Set an arbitrary name for "Partial matrix name", and press the .

The partial matrix with filtered rows is registered to the "Partial matrix name" list.

4. The partial matrix is saved under the following folder path:

• (Folder specified for "Matrix storage folder")*1\Attribute\(whole matrix name folder)\(arbitrarily entered partial matrix

name).csv



21

■ Creating partial matrix by deleting outliers or values other than outliers in the histogram

Operating procedure

1. When the [Histogram] tab is selected, the variable histogram and sample size, quartile (vertical blue line), and a curve

fitted to the Gauss distribution by the least squares method are shown.

2. Sections beyond the whisker tip of the histogram are displayed with a red background. Right-click the histogram and

select [Delete outlier in the box plot.] or [Delete leaving outlier in the box plot.].

Outliers or values other than outliers are deleted from the matrix.

3. Enter an arbitrary name for "Partial matrix name", and press the .

The partial matrix with outliers or values other than outliers deleted in the histogram is registered in the list of "Partial matrix

name".



name).csv


Item Description

(1) Number of bins Specify the number of bins in a histogram (0 to 100).

(2) Display scale Select the vertical axis unit of the histogram between "Freq" and "%".

(1) (2)


■ Creating partial matrix by deleting points in the trend chart

Operating procedure

1. Click the [Trend] tab to show the trend chart.

2. Snap the cursor to the trend chart, then right-click and select [Point deletion] [Delete more than the value of a snap

point all.], [Delete a snap point.], or [Delete below the value of a snap point all.].

3. Enter an arbitrary name for "Partial matrix name", and press the .

The partial matrix with points deleted in the trend chart is registered to the "Partial matrix name" list.



name).csv



21

Correlation analysis by using a correlation graphPerform correlation analysis by displaying a correlation graph of a selected partial matrix.

Operating procedure

1. Select the [Cor. graph] tab.

2. Perform correlation analysis in one of the following four methods:

• Display a correlation matrix. ( Page 236 Correlation matrix display)

• Display a 2-variable scatter chart. ( 2-variable scatter chart display)

• Display a histogram. ( Page 239 Histogram display)

• Display a correlation chart. ( Page 240 Correlation chart display)


■ Correlation matrix display

Operating procedure

1. Select the [Correlation matrix] tab.

Select up to six variables in the "Var." list. A correlation chart (scatter chart) of two variables and the histogram of each

variable are shown.

Item Description

(1) Histogram Display the histogram of each variable.

(2) Scatter chart Displays a correlation chart (scatter chart) of two variables.

The scatter chart shows probability ellipses for 90%, 95%, and 99% (from the inside to the

outside).

(1) (2)


21

■ 2-variable scatter chart display

Operating procedure

1. Click the scatter chart on the [Correlation matrix] tab.

A 2-variable scatter chart is shown on the [Single graph] tab.

The regression line passes through the X minimum and maximum values, that is, the point of contact between

a line drawn in parallel to the Y-axis and the probability ellipse. Therefore, the slope is slightly smaller than the

long axis of the probability ellipse.

2. Click the [Select categolize variable.] button and select a text variable.

3. When a text variable is selected on the "stratification" screen, data linked with the selected text variable is indicated by

the marker.

Item Description

(1) Scatter chart display

screen

Displays a 2-variable scatter chart.

The scatter diagram shows probability ellipses for 90%, 95%, and 99%, and the regression line used to estimate the Y-axis

value from the X-axis value.

(2) Vertical-axis variable chart

display screen

Displays a chart for the vertical axis variable.

(3) Horizontal-axis variable

chart display screen

Displays a chart for the horizontal axis variable.

(1)

(2)

(3)


4. Right-click on the scatter chart display screen, and select [Plot attribute].

5. The attributes of the point snapped by the cursor are shown on the "Plot Attributes" screen. When an item is selected,

plots with the same value are indicated by the marker.

6. Snap the plot with the cursor, right-click it and select [Delete the snapped point.].

7. The plot is deleted and the corresponding data is also deleted from the vertical- and horizontal-axis variable chart display

screens.

When the partial matrix is saved, a partial matrix with the target plot deleted is created.


21

■ Histogram display

Operating procedure

1. Click the histogram in the [Correlation matrix] tab.

The corresponding histogram is shown in the [Single graph] tab.

Item Description

(1) Histogram display screen Displays a histogram.

(2) Statistical quantity Shows the statistics.

(3) Sigma value Shows the sigma values.

(1)

(2) (3)


■ Correlation chart display

Operating procedure

1. Select the [Cor-chart.] tab. Pairs of 2 variables are displayed on the chart from the one with the highest correlation

coefficient first.

A chart with the hazard rate of 5% (Significance level of 5%) and correlation, the background is shown in blue.

2. Right-click on the "Variable name" list and select [Delete variables that have no correlation with other valiables by critical

rate 5%.].

Variables satisfying the condition are deleted from the partial matrix.

3. When a variable is specified in "Reference variable", the specified standard variable and variables are listed in the

descending order of the positive correlation coefficients.

4. Right-click on the "Variable name" list and select [Delete the variable of no correlation by significance level 5%.].

Variables not correlated with the standard variable are deleted from the partial matrix.

5. Selecting [Delete highlight variable.] deletes the highlighted variables from the partial matrix.

6. Clicking the [Undo operation] button cause the partial matrix to revert to the original state before deletion.


21

7. When variable are too many, click the [Delete variable] button while the variables are selected in the "Variable name" list.

Clicking the scatter chart deletes the variables.

8. Click the [Partial matrix operation] button and select [Partial matrix [Saving]].

A partial matrix with the target variables deleted is created.

After operating on a partial matrix, click the [Partial matrix operation] button and select [Partial matrix [Saving]]

to save the operated partial matrix.

9. When the [Histogram] tab is selected, the variable histogram is shown.

10.Selecting the checkbox of "Trend display (Reset graph display)" switches from the histogram display to trend display.


11. Right-click on the "Variable name" list and specify the statistics type from [Row item].

12.By clicking the title row, the statistics displayed in the row can be sorted in ascending or descending order.

13.When the [Cor-coefficient] tab is selected, the basic statistics of each variable, mutual correlation coefficient, and

correlation matrix are shown.

Precautions

• By default, correlation between equipment data and equipment data, and correlation between measurement data and

measurement data are not shown.

• By default, measurement data variables other than that is specified as the standard are not shown. Selecting the checkbox

of "Display the entire correlation." in the [Cor-chart.] tab shows the whole correlation chart including the measurement data.

Item Description

(1) Basic statistics Shows the basic statistics.

(2) Mutual correlation

coefficient

Shows the mutual correlation coefficient.

Up to 100 100 variables are shown.

(3) Correlation matrix Display a correlation matrix.

(1)

(2)

(3)

21 Analyzer21.8 Multiple Regression Analysis (LMR) 243

21

21.8 Multiple Regression Analysis (LMR)Calculates correlation between multiple variables and estimate the variable value using the multiple regression equation.

Operating procedure

1. Select [Advanced analysis] [Multiple regression analysis (MLR)].

2. Perform recurrence calculation.

Page 247 Recurrence calculation

3. Check the analysis chart.

Page 248 Checking analysis chart

24421 Analyzer21.8 Multiple Regression Analysis (LMR)

Screen configurationThe following shows the screen configuration for multiple regression analysis.

Displayed items

Item Description

Whole matrix selecting it on matrix screen Select a whole matrix.

The whole matrix selected in 'Create correlation matrix' is shown by default.


Specify a partial matrix. Select a partial matrix.

A partial matrix of the selected whole matrix can be selected.

[Objective variable] button Click this to set an objective variable and perform recurrence calculation.

[Recurrence calculation] button Click this to perform recurrence calculation with a set objective variable.

[External output] button This button is enabled after a recurrence calculation is performed (while the result of the recurrence calculation is

displayed).

Click this to output the unit space data of the criterion matrix, which is derived in the recurrence calculation

process, in the format available in a programmable controller.

Page 403 External output

The default save file name is shown below. (It can be changed to any name within 255 characters including an

extension.)

LMR_method_param.PMR

For the path of a file to be saved, up to 255 characters (two characters are used for a double-byte character)

including the extension can be specified.

Factor*1 Shows the "Regression variance" or "residual sum of squares".

Degree of freedom*1 Shows the number of samples which can be gathered freely.

• Regression variance: Number of explanatory variables

• Residual sum of squares: n-p-1

p: Number of explanatory variables, n: Number of samples (number of section statistics)

Sum of squares*1 Shows the sum of squares of difference between data and its average value.

• Regression variation: Variation which can be explained with a regression equation.

Sr = i = 1, n(Yi-Ybar)2

Yi: Predicted value, Ybar: Average of measured values

• Residual sum of squares: Variation which cannot be explained with a regression equation.

Se = i = 1, n(yi-Yi)2

yi: Measured value

Average square*1 Shows the value obtained by 'sum of squares' 'Degree of freedom'.

F ratio*1 Shows the value obtained by 'Average square of regression sum of squares' 'average square of residual sum of

squares'.

(1)


21

*1 By clicking the title of each item, the order can be changed in ascending or descending order. However, the objective variable in the first

Item*1 Shows the following items.

• Multiple corr-coefficient: Correlation coefficient between a measured value and a predicted value.

R = Square root of determination coefficient

Ranges from 0 to 1, and a bigger value is better.

• Coefficient of determination: Sum of squares of regression variation Sum of squares of total variation

R2 = "Sum of squares of regression variation (Sr) Sum of squares of total variation (Sr+Se)"

Ranges from 0 to 1, and a bigger value is better.

• Coefficients of determination that have adjusted degree of freedom: Equation which replaces "Sum of squares

of total variation (Sr+Se)" in the equation of the coefficient of determination (R2) with mean square, introducing

the degree of freedom.

R'=(n-1)(1-R2)/(n-p-1)

If the difference between n and p is small, R2 becomes larger just by increasing the explanatory variable p with

no correlation. As a result, R' does not increase simply just by increasing p.

• AIC: nlog(Se)+2(p+1)

Reference for selecting explanatory variables. If the number of explanatory variables is the same, select

variables with the smaller AIC value.

A smaller value is better.

Since AIC is based on the calculation of the maximum log-likelihood that indicates the largest possibility in the

future, it is assumed that the distribution of the estimated value and measured value follows the normal

distribution. AIC is a kind of approximation and its accuracy decreases as the number of explanatory variables

increases. Samples more than twice as many as the explanatory variables are necessary.

Value*1 Shows the value of each item.

(1) Regression

variable list

VIF*1 Variance inflation factor.

If VIF = 1/(1-R2))>10, multicollinearity is suspected.

Single correlation

coefficient*1Single correlation coefficient between an objective variable and explanatory variable.

Single regression

coefficient*1Simple regression coefficient of a single regression equation (objective variable and one explanatory variable).

The first row of the regression coefficient list does not have data.

Partial regression

coefficient*1Regression coefficient of a multiple regression equation.

The first row of the regression coefficient list is a constant term.

Standardized partial

regression

coefficient*1

Regression coefficient with a standardized variable (average = 0, variance = 1).

The constant term in the first row of the regression coefficient list is '0'.

It can be used as reference for contribution to the objective variable.

If correlation between explanatory variables is 0, the standard regression coefficient equals to the single

correlation coefficient.

Standard error*1 Standard deviation of the partial regression coefficient.

Used to estimate sections for the partial regression coefficient.

The first row of the regression coefficient list is a value which can be ignored.

t value*1 Value of "partial regression coefficient standard error". Used for the t-test of the partial regression coefficient.

P-value*1 Possibility of observing statistics more extreme than the t value obtained by the t-test.

LL 95%*1 Section estimated value at the lower limit 95% of the partial regression coefficient.

UL 95%*1 Section estimated value at the upper limit 95% of the partial regression coefficient.

LL 90%*1 Section estimated value at the lower limit 90% of the partial regression coefficient.

UL 90%*1 Section estimated value at the upper limit 90% of the partial regression coefficient.

F value*1 Square of the t-value.

Compared with the F-boundary value to test the partial regression coefficient.

Both sides of

P(F<=f)*1Unused area.

Both sides of F

bound 95%*1Boundary value for section estimate on both sides of F-distribution with a specified degree of freedom.

If the F-value of the partial regression coefficient is greater than this value, "Hypothesis: Influence by partial

regression coefficient is not present" can be rejected.

Both sides of F

bound 90%*1Boundary value for section estimate on both sides of F-distribution with a specified degree of freedom.

If the F-value of the partial regression coefficient is greater than this value, "Hypothesis: Influence by partial

regression coefficient is not present" can be rejected.

Variable mean

value*1Average value per variable.

Var. Sigma*1 Standard deviation per variable.

S_No.*1 Sample number of the partial matrix

Meas.*1 Shows the measured value.

Pred.*1 Shows the predicted value.

Residual error*1 Shows the difference between the measured value and predicted value.

Measured value - Predicted value

Item Description


row of 'Regression variable list' is excluded from the order change.


21

Recurrence calculationPerform recurrence calculation.

Operating procedure

1. Click the [MLR analysis] tab and select the whole matrix and partial matrix created in 'Create correlation matrix'.


2. Select a variable to set as an objective variable in 'Regression variable list'.

3. Click the [Objective variable] button.

When an objective variable is set, recurrence calculation is performed and the result is displayed.

The multiple regression equation is automatically saved in the corresponding partial matrix folder.

• In another way to set objective variables and perform recurrence calculation, right-click in 'Regression

variable list'.

• Recurrence calculation is also performed by selecting the objective variable and clicking the [Recurrence

calculation] button or right-clicking and selecting the [Recurrence calculation] menu.

• When specifying a partial matrix for the primitive 'Regression operation', a recurrence calculation will be

executed at the timing when the Reader reads a CSV file.


• The multiple regression equation for recurrence calculation is as follows:

Predicted value = (Partial regression coefficient of variable n Variable n) + Partial regression coefficient

in the first row (= intercept)

• Only one multiple regression equation is created per partial matrix.

• If any of the following conditions is true, multicollinearity is suspected and the value shown in "VIF" is

marked with "**".

When the sign of the single correlation coefficient (or single regression coefficient) of the objective and

explanatory variables is different from the sign of the partial regression coefficient in the multiple correlation

equation

When "VIF" value = 1/(1-R2)>10

• Multicollinearity occurs when the correlation coefficient is high between explanatory variables. In this case,

recurrence cannot be calculated correctly.

• When multicollinearity is suspected for a variable, the value shown in "VIF" is marked with '**'. If necessary,

delete variables from the partial matrix using 'Create correlation matrix' and perform recurrence calculation

again.


Precautions

• It is necessary to delete unnecessary variables, variables of which values do not change, and outliers from the partial

matrix.

• The number of units of data (number of sections) included in a partial matrix must be three times or more the number of

variables. If it does not satisfy this condition, a recurrence calculation may fail.

• When multicollinearity is detected, an error is indicated. In this case, delete variables which may have high correlation

coefficients from the partial matrix using 'Create correlation matrix' and perform recurrence calculation again.



Checking analysis chartCheck the analysis chart.

Operating procedure

1. Select the [Analysis graph] tab.

2. Right-click on the residual display screen, and select a variable name from [Select variable]. The chart of the specified

variable is shown.

The root mean square (RMSE) can be calculated with the following equation:

a: Measured Value, b: Predicted Value, n: Section count

Item Description

RMSE Shows the root mean square of the errors.

(1) Predicted value/measured value display screen Shows the predicted value/measured value.

(2) Residual display screen Shows the residual.

(3) Scatter chart display screen Displays a scatter chart.

(4) Histogram display screen Displays a histogram.

(5) Statistical quantity Shows the statistics of the variable.

(1)

(2)

(3)

(4) (5)

RMSE = √(Σi=1,n(ai-bi)2) ÷ n


21

3. The scatter chart display screen shows the scatter chart of the predicted and measured values as well as the probability

ellipses (90%, 95%, and 99%).

4. Statistics of specified variables are displayed in 'Statistics'. A histogram, a number of bins of the histogram, and a number

of units of sampled data used for recurrence calculation (number of units of data included in a whole matrix (number of

sections)) are displayed in 'Histogram display' screen.

5. Right-click on the scatter chart display screen, and select [Plot attribute].

6. The "Plot Attributes" screen opens and shows the plot point attributes (state variable values). When an attribute is

selected, plots with the same attribute are indicated by the marker.

If another matrix is selected in "Specify evaluation matrix.", calculation using the multiple regression equation

is performed and the result is shown.

Item Description

(1) Number of bins Specify the number of bins in a histogram (0 to 100).

(2) Display scale Select the vertical axis unit of the histogram between "Freq" and "%".

(1) (2)

25021 Analyzer21.9 Maharanobis - Taguchi method (MT)

21.9 Maharanobis - Taguchi method (MT)Diagnose the variations based on the correlation of variations between multiple variables.

Operating procedure

1. Select [Advanced analysis] [Maharanobis - Taguchi method (MT)].

2. Perform MT distance calculation.

MT distance calculation

3. Check the chart.

Page 253 Checking chart

21 Analyzer21.9 Maharanobis - Taguchi method (MT) 251

21

Screen configurationThe following shows the screen configuration for analyzing by Mahalanobis-Taguchi method.

Displayed items

Item Description

Whole matrix Selects a whole matrix.

The whole matrix selected in 'Create correlation matrix' is shown by default.


Criterion matrix Selects a partial matrix used as the criterion.


Calc-object matrix Selects a partial matrix for calculation.


[Calc MT] button Performs MT distance calculation.

Num. of Ref samples Shows the number of samples (number of units of data included in a whole matrix (number of

sections)) based on MT distance calculation.

(1) Evaluation variable list Shows the evaluation variable list.

(2) Criterion list Shows the criterion list.

(3) MT distance calculation result Shows the MT distance (Mahalanobis distance).

When a multicollinearity error occurs, an error message appears or the result will be 'Nan'.

Display the comment of selected matrix. Shows the history of CSV files read by the selected whole matrix.

Display message of operation. Shows the message for the operation.

[External output] button This button is enabled after performing MT calculation (state in which MT distance is calculated).

Click this to output the unit space data of the criterion matrix, which is derived in the calculation

process of MT distance (Mahalanobis distance), in the format available in a programmable

controller.

Page 410 External output

The default save file name is shown below. (It can be changed to any name within 255 characters

including an extension.)

MT_method_param.PMT

For the path of a file to be saved, up to 255 characters (two characters are used for a double-byte


(1)

(2)

(3)


MT distance calculationPerform MT distance calculation.

Operating procedure

1. Click the [Calc MT] tab, and select the whole matrix, criterion matrix, and calculation target matrix created in 'Correlation

matrix creating'.


For "Criterion matrix" and "Calc-object matrix", the partial matrices of the selected whole matrix are displayed, so select from

among them.

2. When a dynamic characteristic variable is present, select the dynamic characteristic variable from the criterion list.

Selecting a dynamic characteristic variable cancels its selection in the evaluation variable list.

A dynamic characteristic variable is a variable which changes its value according to the input value.

3. Click the [Calc MT] button.

The calculation result of MT distance (Mahalanobis distance) is displayed in 'MT distance calculation result'.

MT distance calculation can also be performed by right-clicking on the evaluation variable list and then

selecting [Calc MT].

Precautions

When an error occurs due to multicollinearity, the error message appears or the calculation result will be 'Nan'.

21 Analyzer21.9 Maharanobis - Taguchi method (MT) 253

21

Checking chartCheck the chart.

Operating procedure

1. Select the [Chart] tab.

2. Right-click on the variable chart display screen, and select a variable name from [Select variable]. The chart of the

selected variable is shown.

Item Description

(1) MT distance (Mahalanobis distance) display screen Shows the MT distance (Mahalanobis distance).

(2) Variable chart display screen Displays a variable chart.

(3) Scatter chart display screen Display a scatter chart.

(4) Variable deviation display screen Shows the variable deviation of a section plotted by cursor.

Num. of Ref samples Shows the number of samples (number of units of data of a whole matrix) based on MT distance

calculation.

(1)

(2) (4)

(3)


3. The scatter chart display screen shows a scatter chart with the vertical axis representing the MT distance (Mahalanobis

distance) and the horizontal axis representing the variable value.

4. The variable deviation of the section plotted on the MT distance (Mahalanobis distance) display screen is shown on the

variable deviation display screen.

5. Right-click on the MT distance (Mahalanobis distance) display screen and choose [Menu] [Variable deflection graph Y

axis fixation].

The vertical axis of the variable deviation display screen is fixed within the range between the maximum and minimum values.

6. Right-click on the MT distance (Mahalanobis distance) display screen or the variable chart display screen and choose

[Plot attribute].

7. The "Plot Attributes" screen opens and shows the plot point attributes (state variable values). When an attribute is

selected, plots with the same attribute are indicated by the marker.

21 Analyzer21.10 Editing Logics 255

21

21.10 Editing LogicsCreate arbitrary analysis logic/diagnostic logic.

A logic is an execution unit of a sequence of processes to perform analysis or diagnosis in Real-time Statistic Diagnosis Tool,

and consists of one or more processes (primitives).

Primitives are executed sequentially in logic units.

A primitive has input and output, similarly to general program functions, and output from an upper-level primitive can be used

as input to a lower-level primitive.

In addition, a variable can be specified to enter a primitive. (Page 255 Available variables for primitives)

Operating procedure

1. Select [Analysis logic - diagnostic logic operation] [Logic editing].

2. Create a logic.

Page 258 Creating logic

3. Execute the logic.

Page 354 Executing logic

Available variables for primitivesThe following table shows the variables available for entering primitives.

(1) Logic

(2) Primitive 1

(3) Primitive 2

(4) Primitive 3

(5) Primitive n

(a) Output from upper-level primitive is used as input to lower-level primitive.

Variable type Description Remarks

Trace variable Numerical value data which varies continuously Variable in a CSV file read by Reader

Text variable Text data which varies in each section Output variable of a primitive or variable in a CSV file

read by Reader

Step variable (state variable) Data which does not vary continuously Output variable of a primitive or variable in a CSV file

read by Reader

Statistical variable (statistics) Statistics of 'average value', 'maximum value', 'minimum

value', 'range', 'standard deviation', 'dispersion', 'median value'

and others.

Output variable of a primitive

Boolean variable (boolean) Value (o or 1) indicates the result of a primitive. Output variable of a primitive

Event variable (event) Time stamp when an event occurs in Real-time Statistic

Diagnosis Tool.

Output variable of a primitive or the start and end

points of an extracted section by Reader.

Logic variable Variable such as a numerical value, text, and conversion table. Variable registered in 'Logic variable setting'.


(2)

(3)

(4)

(5)

…

(1)

(a)

(a)

25621 Analyzer21.10 Editing Logics

Screen configurationThe following shows the screen configuration for editing a logic.

Displayed items

Item Description

Change dataset. Shows dataset selected in 'Management'.

Page 384 Management

The dataset can be changed by clicking it.

Page 257 Changing dataset

[Logic list] tab (1) Logic list Shows the list of logics.

[Logic editing] tab Edits a logic.

Category Select a category of the primitive to create.

Primitive Select a primitive to create.

Logic name Enter a logic name (up to 32 characters).

(2) Logic comment Enter a comment for a logic (up to 1000 characters).

The comment entered here is shown in "Logic comment" in Monitor.

Page 416 Monitor

Model name Do not use this.

[Move highlight step.] button When clicked, the step highlighted in the primitive in the primitive list moves.

While executing the logic manually and the execution result screen is shown, it

changes to the [Run highlight step.] button.

Page 354 Manual execution

(3) Primitives list Shows added primitives.

(3)(1)

(2)


21

Changing datasetChanges the contents of the dataset to display.

Operating procedure

1. For 'Logic editing', click "Change dataset.".

For 'Data display after execution', click "Change dataset.".

The dataset set in 'Management' is shown.

Page 384 Management

2. Set each item to edit the dataset.

Item Description

[DS] tab Dataset (DS) Shows the list of dataset names.

The dataset name is displayed in the format '(unit name)_UNIT'.

The dataset selected in this item is shown on related screens.

[UNIT] tab Unit Shows all units registered in the specified folder.

[Period] tab Period When a unit is selected in "Unit", a data period registered in the unit is shown, one day per one line.

Select the period shown on related screens.

[Sctn] tab Sctn Shows sections in the period selected in "Period".

The section name indicates the start time extracted by Reader, or the start time of the text variable.

When text variables are selected in "Text name" on the [FLT] tab on the "Management" screen, the

text variable values are shown in order from the "1" column.

Page 384 Management

[LOGIC] tab Logic Shows the logic executed during the period selected in "Period" and on the section selected in

"Sctn".

Selecting a logic shows the execution result of that logic.

A logic can be executed for a selected period or section.

[Var.] tab [Trace variable name]

button

Shows variables contained in data registered in the unit.

When a logic is selected in "logic", names of variables output by the logic are also shown.

Statistics When a logic is selected in "logic", names of variables output by the logic are also shown.


Creating logicCreate a logic.

Set and register primitives in an order to be performed.

Precautions

• Register a logic per unit in 'Dataset'. To register a logic, 'Dataset' must be configured in advance.

Page 384 Management

• On the logic list, 'Monitor' is registered by default for all units. 'Monitor' is a logic which monitors the diagnosis results of

'SPC', 'multivariate analysis (single central value format)', and 'guard band'. If 'Monitor' is edited or deleted, the results

cannot be detected in Monitor.

Page 189 SPC, Page 207 Multivariate Analysis (Single Central Value Format), Page 215 Guard Band Diagnosis

• It may not work properly when entering a text directly in the text entry area in the primitive setting. In that case, copy the text

with a text editor or others and paste it in the text entry area.

Operating procedure

1. Select a dataset to register the logic.


2. Select a blank line on the logic list.

3. Select the [Logic editing] tab, and select a primitive or primitives to register by clicking the [Category] and [Primitive]

buttons.

See below for the list of primitives.

Page 261 Primitives

4. Set the parameters for each primitive.

See below for the details of the parameters.

Page 268 Primitive parameters


21

5. "Output" shows a primitive name by default. (A name can be changed as necessary (up to 32 characters).)

When selecting the checkbox of "Save", output from the primitive is added to the dataset variables at the logic execution and

it can be referred in 'Data display after execution'.


Output from a primitive with "Save" checked but is inactive is always added to the variables when the logic is executed.

A comment for the primitive can be changed by changing the content in "Comment" (up to 1000 characters).

When '@@' is specified in 'Output', adding, inserting, or updating a primitive automatically replaces it with a

variable name selected by the primitive (if any).

(Example) When entering "EVENT_EXTRACTION-@@" as an output variable name, selecting the variable

name 'ABC', and adding, inserting, or updating a primitive:

EVENT_EXTRACTION-ABC

6. Right-click on the primitive list and select [Add primitive.] to add the primitive being edited to the end of the primitive list.

If a primitive with the same name already exists in the primitive list, a sequential number (e.g., 01, 02) is automatically

appended to the variable name to avoid duplication of output variable names.

7. Right-click the primitive list while a primitive is selected. Select [Insert primitive.] to insert another primitive on the

currently selected primitive.

8. If a primitive is edited while it is selected, right-clicking it and selecting [Update primitive.] causes the primitive being

edited to be updated with the current contents.

• After editing a primitive, it must be updated, added, or inserted. If none of these operations is done, edited

data will be lost.

• If a primitive is added, inserted, or updated while a period and section are selected in the dataset, the

primitive is run for the selected period and section.


9. The order of primitive steps can be changed by clicking the [Move highlight step.] button while the step to move is

selected.

A primitive is executed sequentially from the primitive of which "STP" is "1".

10. To delete a primitive step, right-click the step and select [Delete step.].

11. Enter a name in "Logic name" (up to 32 characters), and right-click on the primitives list and select [Save logic].

• After editing a logic or primitive, the logic must be saved. If the logic is not saved, edited data will be lost.

• A logic can also be saved by pressing the after setting a logic name for "Logic name".

• If a comment is entered in "Logic comment", the entered comment is shown in "Logic comment" for Monitor.

Page 416 Monitor

12. The logic is saved.

To the logic again, select the logic in the logic list.

13. To delete a logic, select and right-click it in the logic list. Select [Delete logic] to delete the selected logic.


21

PrimitivesThe following shows primitives used to edit logics.Primitive [primitive]

Primitives are classified into the following categories:

Category Description Reference

Event operation Used in the following cases:

• When extracting events at the start time and end time of a subsection to extract a partial

waveform from a waveform collected from the equipment

• When extracting an event to extract time statistics

• When extracting a particular event from multiple events

An event is a time stamp recorded when an event occurs in Real-time Statistic Diagnosis

Tool, and is used as a trigger to control primitives.

For details of the section level, refer to the following:

Page 267 Section level

Page 262 Event operation

Extract section Used in the following cases:

• When extracting a partial waveform to diagnose or analyze a trace variable

• When extracting a calculation section to extract the statistics of an event

• When extracting a subsection from step variable for each value

• When outputting a subsection, the start and end events, and boolean variable

(successful with = 0)

• When deleting a section which is not required for diagnosis or analysis

• When extracting a section for diagnosis or analysis

• When extracting a section for a process for which a subsection is not available such as

power spectrum calculation

• When extracting a single L1 subsection with simple setting



Page 262 Extract section

Constant and variable Used in the following cases:

• When setting the initial value of a variable necessary for logic execution

• When extracting or generating text or step from a step variable, text variable, event,

trace variable, statistics or conversion table (logic variable)

Constant and

variable

Waveform operation Used in the following cases:

• When analyzing data of a trace variable and converting into an applicable form for an

error detection

• When performing a curve fitting for data of a trace variable and extracting a fit waveform

and the coefficient of a fit formula

Page 263 Waveform

operation

Statistic extraction Used in the following cases:

• When extracting statistics from a trace variable in a specified section for diagnosis or

analysis

• When performing the limit pair diagnostics or an SPC by using the extracted statistics

• When extracting event frequency and statistics of the section length enclosed by events

• When extracting statistics from a step variable or file

Page 264 Statistic extraction

Statistic processing Used for statistics extracted by a primitive in the statistics category to perform operation,

save data, and/or change the section level.



Page 264 Statistic

processing

Waveform diagnostics Used to diagnose a waveform using GB (guard band) and extract stabilization points

(points where set upper or lower limit width was entered) from a waveform.

Page 265 Waveform

diagnostics

Statistic diagnostics Used to diagnosis by using statistics extracted in a primitive. Page 265 Statistic

diagnostics

Decision logic Used to control step execution of primitives. Page 265 Decision logic

Output Used to report an error and output the error log to a file when the error is detected during

logic execution.

Page 266 Output

Special usage Used to define a primitive which is not usually used or does not require parameter setting. Page 266 Special usage


■ Event operationPrimitives classified in this category are listed below.

■ Extract sectionPrimitives classified in this category are listed below.

Primitive Description Parameter settings

Event extraction Extracts an event by specifying one of the following methods:

• Text matching

• Offset from section start time

• Threshold of trace variable

• Maximum/minimum value of trace variable

Page 268 Event

extraction

Step change event extraction Detects change of a step variable and output the time of change as an event. Page 270 Step change

event extraction

Variable threshold event

extraction

Uses a threshold to detect events and peaks from trace variable. Page 271 Variable

threshold event extraction

Event extraction at regular

intervals

Extracts lower-level sections divided equally or at an even interval from a section. Page 272 Event

extraction at regular intervals

Event consolidating & shift Integrates several different events and converts them into a single event. Shifts the time of

the converted event by a specified amount of time.

The output variable name of this primitive is used as the name of the integrated event.

Page 273 Event

consolidating & shift

Order specified event extraction Extracts an event from several events according to a specified order. Page 274 Order

specified event extraction

Event filter Filters events based on a specified variable value.

Outputs events passing through the filter only.

Page 275 Event filter

Text change event extraction Detects change of text and output the time of change as an event. Page 276 Text change

event extraction

Statistic event extraction Divides a section into subsections of a specified length, and extracts specified statistic

data per variable per subsection.

If the statistics data does not update the maximum or minimum value of a subsection

consecutively for the specified number of subsections, the start time event for the first

subsection is extracted.

Page 277 Statistic event

extraction


Extract subsection Extracts subsections (L1, L2, and L3).

Three configuration modes are available: simplified setting, detailed setting, and state

variable.

If the upper-level section is not extracted, the lower-level section cannot be extracted.

The range of subsections is specified by an event or change of the state variable.

Page 278 Extract

subsection

Specified range extraction Extracts a section defined by the start and end events.

Or, deletes a section defined by the start and end events and outputs the rest.

Page 282 Specified

range extraction

L-1-Section extraction Extracts a single L1 section as specified by the start and end conditions.

When an L1 section has been already extracted, the succeeding sections are extracted.

Page 283 L-1-Section

extraction


21

■ Constant and variablePrimitives classified in this category are listed below.

■ Waveform operationPrimitives classified in this category are listed below.


Constant generation Sets a constant as the initial value. Page 286 Constant

generation

Reading numeric LOGIC variable Outputs a value set to numeric by a logic variable as statistics data. Page 288 Reading

numeric LOGIC variable

Text variable connection Connects multiple texts and outputs the result as a single text variable.

The time to extract texts can be specified.

Page 289 Text variable

connection

Numeric variable extraction Extracts the value at the time specified by an event as a step variable from a step

variable, trace variable, and statistics.

An offset (in seconds) can be set for an event.

When the offset time is out of the specified section, no value is extracted.

Page 290 Numeric

variable extraction

Execution conversion Outputs values matching the input condition using a conversion table registered by a logic

variable.

Page 291 Execution

conversion

Step variable filter Compares two state variables or a state variable and a numeric value and converts the

time zone when they match into a value.

Page 292 Step variable

filter


Waveform conversion Perform one of the following:

• Moving average (simple moving average, polynomial fit, median, smoothing by MAD)

• Waveform differentiation (n-order differential)

• Waveform integration (n-order integral)

• Waveform difference (difference by shifting own waveform)

• Auto-scaling (converts to average value = 0, standard deviation = 1)

• Range-scaling (converts amplitude to 1)

• Centering (shifts Y value to make average value =0)

• Wave statistic operation (calculation between waveforms or waveform and statistics)

• Logarithmic conversion (common, natural, logarithmic conversion with base = 2)

Page 293 Waveform

conversion

Waveform interpolation Interpolates trace variable and converts it into a short sampling interval. Page 297 Waveform

interpolation

Shift in X axis Shifts a variable in the time axis direction. Page 298 Shift in X axis

Max. Avg. Min. waveform ext. Extracts the following from selected multiple waveforms:

• Maximum value waveform extracted only the maximum value

• Minimum value waveform extracted only the minimum value

• Average value waveform extracted only the average value

For the maximum or minimum value waveform, a text variable which indicates the trace

variable name for the maximum or minimum values of the points on the waveform.

Page 299 Max. Avg.

Min. waveform ext.

Linear curve fit Performs curve-fitting using a linear equation and outputs a waveform and coefficient. Page 300 Linear curve

fit

Polynomial curve fit Performs curve-fitting using a polynomial equation and outputs a waveform and

coefficient.

Page 301 Polynomial

curve fit

Exponential curve fit Performs curve-fitting using an exponential equation and outputs a waveform and

coefficient.

Page 302 Exponential

curve fit


■ Statistic extractionPrimitives classified in this category are listed below.

■ Statistic processingPrimitives classified in this category are listed below.


Individual trace statistic Extracts specified statistics from trace variable. Page 303 Individual

trace statistic

Event specified trace value ext. Extracts the value at the time when an event occurs from multiple trace variable data. Page 304 Event

specified trace value ext.

Multi trace statistic Extracts multiple statistics data items from multiple trace variable data.

The output statistics name is "output variable name" + "trace variable name" + "statistics

type".

To shorten the name, change the output name to "_" (underscore).

Page 305 Multi trace

statistic

Area statistic Calculates time for which trace variable exists within a range defined by specified upper

and lower limits.

Page 307 Area statistic

Event statistic Performs the following per specified section:

• Event section: Extracts the section length as statistics.

• Event frequency: Extracts the event occurrence frequency as statistics.

• Event interval: Extracts the event occurrence interval as statistics.

Page 308 Event statistic

Range specified event statistic Extracts multiple sections defined by start and end events and calculates the statistics of

the section length.

Page 309 Range

specified event statistic

Event specified variable value ext. Extracts the Y value of the variable at the time when the specified event first occurred

during the specified section.

Page 310 Event

specified variable value ext.


Multi-section total Calculates statistics from the current and past section data.

To reference output from a primitive after the current step, run the primitive in the FD

mode (error detection mode). It cannot be executed in the analysis mode.

When [Reset at Zero Value] is checked, the aggregated value can be reset each time the

value for the current section becomes "0".

Page 311 Multi-section

total

Subordinate section total Performs a specified operation on the statistics of a lower-level section and outputs the

result as the statistics of the upper-level section.

Page 312 Subordinate

section total

Multi-section calculation Performs an operation on statistics of the past and current sections. Page 313 Multi-section

calculation

Calculation in present section Performs an operation on the current sections or extracts a central value.

When a multi-operation form is set, a polynomial operation is performed on multiple

variables concurrently.

Page 314 Calculation in

present section

Raises level to section statistic. Upgrades statistics of a subsection to section statistics of a specified section number. Page 315 Raises level

to section statistic.

Statistic level down Downgrades statistics extracted on an upper level by one level so that it can be

referenced on the lower level.

Page 316 Statistic level

down

Regression operation Performs multiple regression analysis (LMR) calculation by specifying a partial matrix and

outputs a predicted value.

Page 317 Regression

operation

MVA statistic Outputs MT distance, Q statistic (SPE), and Euclidean distance (EU distance) by

specifying a partial matrix.

By specifying a text variable, calculation is performed for each text value in a partial

matrix.

Page 318 MVA statistic


21

■ Waveform diagnosticsPrimitives classified in this category are listed below.

■ Statistic diagnosticsPrimitives classified in this category are listed below.

■ Decision logicPrimitives classified in this category are listed below.


Enhanced trace GB Performs the guard band diagnosis of the waveform for each section or L1 subsection.

The guard band is created from the reference waveform or registered directly.

When a value exceeds the guard band consecutively for the number of excess points,

'Boolean variable = 1' is output.


Section level

Page 320 Enhanced

trace GB

Stability deviating GB Detects stabilization points and points deviating from stabilization consecutively.

Outputs the event at the stabilization deviation time, time to stabilize, and stabilization

time.

Page 322 Stability

deviating GB

Variable limit pair GB Diagnoses the upper and lower limit values of a waveform.

A value equal to the upper or lower limit is regarded to be excessive.



Page 324 Variable limit

pair GB

Amplitude change detection Sections the moving window horizontally at the specified point.

Calculates the average amplitude of the waveform in each window.

The average amplitude of the waveform in the first window is assumed to be 100%.

When the average amplitude of the waveform in the second window becomes larger or

smaller than that of the first window by the specified percentage, it is detected as an

amplitude change point.

Page 325 Amplitude

change detection

Slope change detect. Sections the moving window at the specified point.

Performs liner fitting on the waveform in each window to calculate the slope.

Detects the point where the slope increase or decreases as well as the peak. A repeated

detection count can be specified.

Page 327 Slope change

detection


SPC diagnostics Performs SPC diagnosis. Page 330 SPC

diagnostics

Limit pair diagnostics Diagnoses multiple statistics data pieces with the same upper and lower limits.

When a conversion table is used, multiple statistics data pieces are diagnosed with the

same upper and lower limits set by a logic variable, and the diagnosis result is inserted

into the message template.

Page 331 Limit pair

diagnostics

Moving upper/lower limit diag. Extracts statistics from specified moving sections.

Diagnoses the extracts statistics.

Page 333 Moving upper/

lower limit diagnostics


Boolean operation Performs a boolean operation on boolean variables.

Detects the boolean variable and terminates a logic.

Page 334 Boolean

operation

Boolean level change Upgrades a boolean value of a subsection of the specified subsection number to a section

boolean value.



Page 335 Boolean level

change

Judge by text Determine the value of a text variable value and output a boolean variable.

By selecting the checkbox of "Logic closing option", the logic will be ended depending on

a boolean variable value.

Page 336 Judge by text

Judge by step Determine the value of a step variable value and output a boolean variable.

By selecting the checkbox of "Logic closing option", the logic will be ended depending on

a boolean variable value.

Page 337 Judge by step

IF When the boolean variable value is "1", the steps between IF and ELSE are executed. Page 338 IF

ELSE Steps following the IF to ELSE section are executed when the boolean variable value is

"0".

Page 339 ELSE

ENDIF Terminates the IF ELSE statement. Page 339 ENDIF

CASE Executes the CASE statement.

When the input variable value and comparison value match, the variable value of the

specified type is output.

Page 339 CASE


■ OutputPrimitives classified in this category are listed below.

■ Special usagePrimitives classified in this category are listed below.


Save CSV data Saves the waveform and statistics in the CSV format.

For the unit of one file, specify the unit of the period to store the file.

For the write option, specify whether to overwrite the entire file for each saving or append

new sections to the existing file.

When [[EQ]], [[FLT]], [[SEQ]], [[START]], [[END]], and [[variable name]] are added to the

file name, an appropriate value is inserted in each [[]].

This allows for saving CSV data from the same period in different files with different file

names.

Page 340 Save CSV

data

Read & insert template Reads the template and inserts the specified text. Page 342 Read & insert

template

Write log and error notification Displays a pop-up for an error.

Writes log data to display in the operating screen of 'Diagnostic logic result display'.

Page 382 Displaying a Diagnostic Logic Result

A response data file can be output by using this in combination with a waveform

diagnostic logic not having a function to display a pop-up window or a statistic diagnostic

logic.

Page 344 Write log and

error notification

Write template Outputs the template to the specified folder. Page 345 Write template


Moving max min Generates a moving maximum or minimum value waveform. Page 346 Moving max

min

Parametrical conversion Converts a single input value into multiple output values using the I/O conversion table.

Interpolation is performed linearly.

Page 347 Parametrical

conversion

Step variable generation When specifying an event, assigns a numeric value to the event and converts it to a state

variable.

When specifying time, specifies a numeric value and time to retain the value and creates

a state variable.

Page 348 Step variable

generation

Boolean judgment statistic saving Saves a specified value according to a boolean variable value.

For the automatic specification value update SPC, a diagnosis value is saved when the

result of SPC determination is normal and no value is saved when the result is not normal.

Page 350 Boolean

judgment statistic saving

Read measurement data. Reads the measurement data.

Reads a value which matches the key item value from a CSV file as statistics.

Up to two key items are ANDed for judging whether they match each value.

Reads CSV data from multiple sections of a single CSV file or from multiple CSV files.

Page 351 Read

measurement data

Simple diagnosis Performs diagnosis registered on the operation screen of 'SPC', 'Multivariate analysis

(single central value format)', or 'Guard band diagnosis'.

Page 189 SPC



This primitive is defined in the logic "Monitor" that is provided by default.

This primitive cannot be added, changed, and/or deleted.

Frequency filter Performs frequency filtering. Page 352 Frequency

filter

Power spectrum Calculates the power spectrum. Page 353 Power

spectrum


21

■ Section levelJust as Reader extracts an expansion section from a unit section, a subsection can be extracted from a unit section or

expansion section.

Subsections has levels of L1 (upper) to L3 (lower), and the subsection of L1 is defined as L2, and the subsection of L2 is

defined as L3.

Multiple subsections can be extracted on the same level and are assigned ascending subsection numbers 1, 2, 3, ...

chronologically from the oldest subsection.

(1) Unit section or expansion section

(2) L1 subsection (Numbers 1, 2, ... n are assigned to the 1st to the nth sections.)



(a) 1st section

(b) 2nd section

(c) nth section

(1)

(2)

(3)

(4)

(2)

(3)

(4)

(2)

(3)

(4)

(a) (b) (c)


Primitive parametersSetting items (parameters) for each primitive are explained below.



■ Event extractionExtracts events according to the selected event type (text, time, threshold, maximum value, or minimum value).

Item Description Output

Sctn Specifies the level of sections to output.

Text Extracts events of the time when the text variable

changes to the specified value.

• Extracts all events satisfying the

condition.

• When the comparison operator is

[Any], an event is extracted when

ever text changes.

[Text variable name] button*1 Selects a text variable used for comparison to extract

events.

Operator*1 Select one of the following options:

• =: Equal to

• !=: Not equal to

• Any: Value changed

Comparison text*1 When setting a comparison operator other than "Any",

set a text to compare (up to 32 characters).

A wild card ('*' for one or more characters and '?' for

one character) can be set.

Include section head time*1 When checked, the value of the section or subsection

start time is used for comparison.

Time Extracts events according to the offset from the

section start time.

Extracts one event satisfying the

condition.

Offset value*2 Set an offset value (0.0000 to 2147483647.0000)

from the start time of a section or subsection.

Offset unit*2 Select "sec" or "%".

• "%" refers to a percentage of the upper-level

section.


21

*1 It can be set when "Text" is selected.*2 It can be set when "Time" is selected.*3 It can be set when "Threshold" is selected.*4 It can be set when "Max min" is selected.

Threshold Extracts events from trace variables using a fixed

threshold.

Extracts all events at the times when

the threshold is crossed.

[Trace variable] button*3 Select a trace variable.

Threshold*3 Enter a threshold value (-2147483648.0000 to

2147483647.0000).

Threshold unit*3 Select "sec" or "%".

• "%" represents the percentage of the threshold with

the minimum value and maximum value of the trace

variable within the section regarded as 0% and

100%, respectively.

Direction of detection*3 Select one of the following options:

• Up: An event is extracted when a waveform rises.

• Down: An event is extracted when a waveform falls.

Specify immediately before.*3 • By selecting this, the time of the former point is

extracted as an event when crossing a threshold

between two points.

• By unselecting this, the time of the latter point is

extracted as an event.

Max min Extracts an event at the time when the maximum or

minimum value occurs.

Extracts an event at the time when

the maximum or minimum value

occurs.[Trace variable] button*4 Select a trace variable.

Extraction point*4 Select "Max" or "Min".

• Max: An event is extracted at the time when the

trace variable data becomes the maximum during

the specified section.

• Min: An event is extracted at the time when the

trace variable data becomes the minimum during

the specified section.

When there are many points which take the maximum

value or minimum value, the time of the first point is

extracted as an event.

Item Description Output


■ Step change event extractionExtracts events of the time when values before and after the step variable changes to the specified value.

Setting item Description Output

Sctn Specifies the level of sections to output. All events on the time which

matched to a specified condition are

extracted by detecting the variation

of a step variable.

Variable that extracts event Select a step variable to extract events.

Numerical value

before it changes

Select one of the following options:


• =: Equal to

• >=: Equal to or greater than

• <=: Equal to or less than

• >: Greater than

• <: Less than


When "Any" is selected, all change points are

targeted regardless of values changed.

Comparison value Specify a value unless "Any" is selected in "Numerical

value before it changes".

Enter a value (-2147483648.0000 to

2147483647.0000) or select a variable (step variable,

statistics, or logic variable) from the menu.

Numerical value after

it changes



• =: Equal to



• >: Greater than

• <: Less than


When "Any" is selected, all change points are

targeted regardless of values changed.

Comparison value Specify a value unless "Any" is selected in "Numerical

value after it changes".

Enter a value (-2147483648.0000 to

2147483647.0000) or select a variable (step variable,

statistics, or logic variable) from the menu.

Include section head time When checked, the value of the section or subsection

start time is used for comparison.

Shift time Shifts the output event forward or backward.

Enter time (-2147483648.0000 to 2147483647.0000

seconds) or select a variable (statistic, logic variable,

or step variable).


21

■ Variable threshold event extractionExtracts events by specifying a threshold or peak.


Sctn Specifies the level of sections to input and output. All events on the time which matched to a specified condition

are extractedVariable that extracts

event

Select a trace variable name to extract events.

Threshold Enter a value (-2147483648.0000 to 2147483647.0000) or select a

variable (statistics, or logic variable).

Extracts an event in a range exceeding a threshold.

Threshold of the upper-level section can be referenced.

Threshold unit Select one of the following options:

• Value: Uses the trace variable data as the threshold

• %: Uses the percentage of trace variable amplitude within the

specified section as the threshold, with the maximum amplitude

regarded as 100%

Detection type Select one of the following options:

• Up: An event is extracted when a value matches or exceeds a

threshold at the timing of the rising of a waveform.

• Down: An event is extracted when a value matches or falls below

a threshold at the timing of the falling of a waveform.

• Peak (+): An event is extracted when a peak in the positive

direction exceeds a threshold is detected.

• Peak (-): An event is extracted when a peak in the negative

direction falls below a threshold is detected.

Width of peak

detection

Allows for detecting peaks without being affected by a noise.

Specify the minimum value possible.

Detection start time

specification event

Specify the time to start detection.

If not specified, detection starts at the lead time of the section level.

Filter Select from "All", "Max",or "Min".

• Max: Detects the maximum peak Y value from multiple peaks.

• Min: Detects the minimum peak Y value from multiple peaks.

Specify event

generation order

When "All" is selected for "Filtering" and multiple events occur, one

of them is extracted according to the specified order (0 to

2147483647).

If set to '0', all events are extracted.


■ Event extraction at regular intervalsExtracts events at a constant interval.


Sctn Specifies the level of sections to input and output. • Extracts events at a specified

interval.

• When "Equal intervals" is

specified in "Extraction type",

length of the last section is shorter

than the specified time interval

unless the entire section can be

divided by the specified time

interval.

(1) Extraction type Select one of the following options:

• Equal dividing: Extracts events which divide an

section at a regular interval. In this case, specify the

number of divisions (2 to 10000) and the number of

events (0 to 10000).

• Equal intervals: Extracts events at a specified

interval. In this case, specify an interval (0.0000 to

2147483647.0000) and the number of events (0 to

10000).

If the number of events is set to '0', all events are

extracted.

Output section head event. If checked, the event at the section start time is also

included in the extracted events.

Output section end event. If checked, the event at the section end time is also

included in the extracted events.

Cycle event start offset Specify an offset from the time to start event

extraction (start time of the specified section).

A fixed value (0.0000 to 2147483647.0000) or

variable can be specified as an offset. If both are

specified, the variable is used.

Cycle event end offset Specify an offset (-2147483648.0000 to 0.0000) for

the time to end an event extraction (end time of a

specified section).

(1)


21

■ Event consolidating & shiftIntegrates several different events and converts them into a single event.


Event name Selects events to aggregate. • Integrates several events with different names and converts

them into a single event.

• Shifts the converted event forward or backward. Note that an

event which was shifted out of the section is not output.

Shift time Shifts the aggregated event.

Enter time (-2147483648.0000 to 2147483647.0000 seconds) or

select a variable (statistic, logic variable, or step variable).

If only one event is selected, this event is shifted.

Consolidate into one

event when multiple

events are generated

at the same time.

Specify whether to aggregate multiple events into a single event or

keep them all if they occur at the same time.

When a subsection is extracted using multiple events occurring at

the same time as the start and end events, a subsection of which

length is "0" is extracted.


■ Order specified event extractionExtracts an event from several events with the same name according to a specified order.


Sctn Specifies the level of sections to output. Extracts an event from several events according to a specified

order.Event extracted

specifying order

Select an event to extract.

Specify event

generation order

Specify the occurrence order (0 to 2147483647) of events to be

extracted from the start time of a specified section.

If the order is set to "0" or a value larger than the number of events

present in the section, the last event is selected. Overlapping events

are counted individually.


21

■ Event filterFilters events based on a specified variable value.


Sctn Specifies the level of sections to input and output. Outputs all events satisfying the condition during the specified

section.Filtered event Specify an even to be filtered.

Filter variable

(Statistics, Step

variable name, and

Text variable name)

Specify the statistical variable name, step variable name, and text

variable name.

Comparison value

(Comparison statistic,

Comparison

numerical value, and

Comparison text)

Enter the fixed statistics (-2147483648.0000 to 2147483647.0000),

fixed numeric value (-2147483648.0000 to 2147483647.0000), and

fixed text (up to 32 characters).

AND/AND2 When two or more filter variables are specified, they are ANDed.

Operator For a numeric value, select from "=]", ">=", "<=", ">", "<", or "!=".

For text, select "=" or "!=".

• =: Equal to



• >: Greater than

• <: Less than



■ Text change event extractionExtracts events of the time when values before and after the text variable changes to the specified value.


Sctn Specifies the level of sections to output. All events on the time which matched to a specified

condition are extracted by detecting the variation of

a text variable.Variable that extracts event Specify the text variable to extract events.

Text before

changing

Select "Any", "=", or "!=" as the comparison operator. When

"Any" is selected, all change points are targeted regardless

of values changed.


• =: Equal to


(1) Comparison text Specify a value unless "Any" is selected in "Text before

changing".

Enter a text (up to 32 characters) or select a variable (text

variable or logic variable) from the menu.

A wild card character ('*' represents one or more characters

and '?' represents a single character) can be used.

Text after changing Select "Any", "=", or "!=" as the comparison operator.

When "Any" is selected, all change points are targeted

regardless of values changed.


• =: Equal to


(2) Comparison text Specify a value unless "Any" is selected in "Text after

changing".

Enter a text (up to 32 characters) or select a variable (text

variable or logic variable) from the menu.

A wild card character ('*' represents one or more characters

and '?' represents a single character) can be used.

Include section head time When checked, comparison starts from the section or

subsection start time.

Shift time Shifts the output event forward or backward.

Enter time (-2147483648.0000 to 2147483647.0000

seconds) or select a variable (statistic, logic variable, or

step variable).

(1)

(2)


21

■ Statistic event extractionDivides a waveform into subsections and extracts the specified statistics.

If the specified statistics data does not update the maximum or minimum value consecutively for the specified number of

subsections, the event for the first subsection is extracted.


Sctn Specify the level of a section to be output. • Extracts an event at the subsection start time and the

statistical value at that time.

• When extracting, 'Boolean variable = 1' is output. When not

extracting, 'Boolean variable = 0' is output.

Diagnostics variable Specify the target waveform.

Number of subsection

length points

Specify the length of a subsection when dividing a section into

subsections by using a number of points (1 to 2147483647).

Extraction statistic

type

Specify the statistics extracted from the waveform.

Number of detection

judgment subsections

Specify the number (2 to 2147483647) of subsections to detect the

maximum value, minimum value, and stabilization (that is a state in

which values are in a specified range).

Extraction option Select one of the following options:

• Stability extraction: Events of the stabilization of statistics are

extracted.

• Minimum value extraction: When the minimum value in statistics

is not updated consecutively for the specified number of

subsections, extract the event at the start time of the subsection

containing the minimum value.

• Maximum value extraction: When the maximum value in statistics

is not updated consecutively for the specified number of

subsections, extract the event at the start time of the subsection

containing the maximum value.

Stability detection

guard band width

Shown when "Stability extraction" is selected in "Extraction option".

Enter the upper and lower limit widths (0.0000 to

2147483647.0000).

Extract including

equal

Shown when "Minimum value extraction" or "Maximum value

extraction" is selected in "Extraction option".

Extracts an event even if the statistics does not change.


■ Extract subsection

• Setting items when selecting "Simple setting" for "Set mode"

Extracts a section enclosed by two successive events as a subsection.

Precautions

When setting "Simple setting" for "Set mode", if there is any item selected when setting "Advanced" for "Set mode", the

selected contents are applied.


Set mod Select from "Simple setting", "Advanced", or "Step variable".


Subsection extraction

start condition

Select an event, step variable, or text variable. • Extracts a section enclosed by the subsection extraction start

and end conditions as an L1 section.

• Step and text variables are used assuming that a change

point is an event.

• The following are output:

Boolean variable (Successful: 0, failed: 1)

Event (Start and end of subsection extractions)

Statistics (Number of subsections extracted)

Subsection extraction

end condition

Select an event, step variable, or text variable.


21

• Setting items when selecting "Advanced" for "Set mode"

Extracts a section enclosed by two events as a subsection by specifying a condition.


Extract level Specify the section level to extract events. • When the same event is specified for both of the start and

end events or a step variable is specified, consecutive

sections are extracted.

• Sections (L1, L2, and L3), boolean variable (successful = 0,

failed = 1), and statistics (number of sections extracted in the

same section level in a logic) are extracted.

• When extracting a dummy section, 'Boolean variable = 1' is

output.

Higher subsection No. Specify the order of the upper-level subsection (0

to 65535) to extract subsections.

When "0" is specified, all subsections are targeted

for extraction.

When a number greater than the maximum upper-

level section number is specified, the last section

is targeted for extraction.

Extract subsection No. Specify the order of subsections to be extracted (0

to 65535).

When "0" is specified, subsections are extracted

repeatedly (repeat extraction).

When a number greater than the maximum upper-

level section number is specified, the last section

is extracted.

Override Check to overwrite already extracted sections at

the same level.

When a subsection is overwritten, data extracted

for that subsection is deleted.

Range of extraction Specify the range of lengths of the subsection to

extract.

Subsections which are shorter or longer than the

specified range are not extracted.

• When "%" is set for "Extraction range unit",

specify this in the range from 0.0000 to

100.0000.

• When "sec" is set for "Extraction range unit",

specify this in the range from -

2147483648.0000 to 2147483647.0000.



Extraction range unit Specify the unit of the extraction range (length) in

"%" or "sec".

• "%" refers to a percentage of the upper-level

section.

• When the same event is specified for both of the start and

end events or a step variable is specified, consecutive

sections are extracted.

• Sections (L1, L2, and L3), boolean variable (successful = 0,

failed = 1), and statistics (number of sections extracted in the

same section level in a logic) are extracted.

• When extracting a dummy section, 'Boolean variable = 1' is

output.

Subsection

extraction start

condition


Set offset at start

time.

Specify an offset (-2147483648.0000 to

2147483647.0000) from the start event to start a

section.

A logic variable can also be selected.

When offsetting backward, specify a negative

value. When offsetting forward, specify a positive

value.

More than 2 start

candidates.

Specify either "First" or "Last" to determine which

one of multiple start events occurring

consecutively should be used.

Subsection

extraction end

condition


Set offset to end

time.

Specify an offset (-2147483648.0000 to

2147483647.0000) from the end event to end a

section.

A logic variable can also be selected.

When offsetting backward, specify a negative

value. When offsetting forward, specify a positive

value.

More than 2 end

candidates.

Specifies which one of multiple end events

occurring consecutively should be used or

whether the first or last end event should be used.

Option when section cannot be

extracted

When no section could be extracted, a dummy

section is extracted.

• No dummy extraction.: Nothing is performed.

• Dummy extraction: A dummy section with the

specified section length is generated from the

end time of the previous section or from the

section start time.

• Quit the logic: The logic terminates.

Dummy section length (second) Specify the length of a dummy section (0.0000 to

2147483647.0000).

When '0' is specified, a zero-length section is

extracted.


21

If a subsection is not extracted at an expected location, the subsection number may differ from an expected

number. A dummy section is a temporary section to prevent this situation (i.e., to align the subsection

numbers.

• Setting items when selecting "Step variable" for "Set mode"

Extracts a subsection with the subsection number same as the variable value.


Extract level Specify the section level to extract events. • As the step variable changes its value, a subsection with the

subsection number that is the same as the variable value is

extracted.

• When the step variable values change from '1' to '3' and to

'5', skipping '2' and '4', the length of the subsections No.2

and No.4 is zero.

• When the step variable values change from '1' to '3' and then

to '2', not in the ascending order, the subsections No.1 and

No.3 are extracted.

• The following are output:

Boolean variable (Successful: 0, failed: 1)

Event (Start and end of subsection extractions)

Statistics (Number of subsections extracted)

Disregard value, and

extract subsection

only by change.

When checked, extracts subsections by connecting points where

the step variable changes its value.

Subsections are extracted regardless of the ascending or

descending order as well as the direction.

Step variable Select a step variable.


■ Specified range extractionDelete a section between events of a specified variable or retain the section and extract the section data.


Start event Specifies the section start event. • When "Deletion b/w Ev." is selected in "Specify extraction or

deletion.", the specified section is deleted and section data is

output.

• When "Extraction b/w Ev." is selected in "Specify extraction

or deletion.", the specified section is left and section data is

output.

• An extracted waveform is forward-aligned from the beginning

of the original section.

• When an extraction is successful, 'Boolean variable = 0' is

output. When an extraction fails, 'Boolean variable = 1' is

output.

End event Specifies the section end event.

Specify extraction or

deletion.

Select "Deletion b/w Ev." or "Extraction b/w Ev.".

• Deletion b/w Ev.: The specified sections between events is

deleted and the section data is output

• Extraction b/w Ev.: The section data is output with the specified

sections between events left

Variable name Select a trace variable, event, or step variable.


21

■ L-1-Section extractionSpecify the start and end conditions to extract the subsection of L1.


Step Do not use this. ("1" can only be selected.) A subsection and an event of the start and end time are output.

When an extraction is successful, 'Boolean variable = 0' is

output. When an extraction fails, 'Boolean variable = 1' is

output.

Specify the type of

start condition.

Select from "Start time", "End time", "Event", "Text", "Step", or

"Threshold".

Offset (sec) Specify an offset (-2147483648.0000 to 2147483647.0000) from the

specified start condition.

When offsetting backward, specify a negative value. When

offsetting forward, specify a positive value.

Offset from mid point When selecting this, specify a value for "Offset (sec)" in case where

the midpoint of a subsection to be extracted is set as 0. For "Offset

(sec)", specify a negative value to offset backward from the midpoint

or a positive value to offset forward from the midpoint.

Extracts an L-1 section which spans from the middle point of the

section by the forward and backward offset lengths.

Specify type of end

condition.

Select from "Start time", "End time", "Event", "Text", "Step",

"Threshold", or "Start Subsection".


specified end condition.



Min section length

(sec)

Specify the minimum section length (-2147483648.0000 to

2147483647.0000).

Sections of which length is longer than the specified number of

seconds are extracted.

Logic closing option Option used when executing the logic by creating temporary

sections per CSV file.

When checked, one of the following can be selected:

• End with all subsections failure: When no subsection is extracted,

the step in this primitive terminates the logic.

• End it if there is a failure subsection: If any of subsections is not

extracted, the step in this primitive terminates the logic.

When sections are extracted and the logic finishes normally, the

next temporary section is not generated and the logic is not

executed.

To use this function, "Execute logic by file unit before ending

section" under 'Extended section definition' must be checked in

Reader.

Page 155 Defining an extended section


• Setting items when selecting "Event" for "Starting condition type" or "End condition type"

• Setting items when selecting "Text" for "Starting condition type" or "End condition type"

• Setting items when selecting "Step" for "Starting condition type" or "End condition type"


Variable name Select an event for the start condition or end condition. • Outputs the subsection and events and the start and end

times.

• When the extraction of a subsection is successful, 'Boolean

variable = 0' is output. When the extraction fails, 'Boolean

variable = 1' is output.


Variable name Select a text variable for the start condition or end condition. • Outputs the subsection and events and the start and end

times.




Operator Select "=" or "!=".

Comparison text Set a text to compare (up to 32 characters).


Variable name Select a step variable for the start condition or end condition. • Outputs the subsection and events and the start and end

times.




Operator Select "=", "!=", "<", ">", "<=", or ">=".

Comparison value Enter a numerical value to compare (0.0000 to 2147483647.0000).


21

• Setting items when selecting "Threshold" for "Starting condition type" or "End condition type"


[Trace variable]

button

Select a trace variable for the start condition or end condition. • Outputs the subsection and events and the start and end

times.




Threshold/Variable

name

Enter a threshold value (-2147483648.0000 to 2147483648.0000)

or select a step variable for the start condition or end condition.

Direction of detection Select one of the following options:

• Up: A value exceeds a threshold.

• Down: A value falls into a threshold.

• =: A value becomes the same as a threshold.

Threshold unit Select "Value or "%" ('%' for the waveform amplitude).


■ Constant generationSets numeric constants.

• Setting items when selecting "Statistic" for "Type"

• Setting items when selecting "Boolean" for "Type"


Type Select the output constant type from the following options:

"Statistic", "Boolean", "Step variable", or "Text variable".

• Statistic: Numeric value

• Boolean: 1, 0

• Step variable: Numeric value

• Text variable: Text

Outputs a constant of the specified type.


Sctn Select the section level. Outputs a constant of the specified type.

Input numerical value. Enter a numerical value (-2147483648.0000 to 2147483647.0000).


Sctn Select the section level. Outputs a constant of the specified type.

Select ON/OFF of

execution.


• ON: Output variable = 1

• OFF: Output variable = 0


21

• Setting items when selecting "Step variable" for "Type"

• Setting items when selecting "Text variable" for "Type"


Input numerical value. Enter a numerical value (-2147483648.0000 to 2147483647.0000). Outputs a constant of the specified type.


Input character. Enter a character string (up to 32 characters). Outputs a constant of the specified type.


■ Reading numeric LOGIC variableReads the step variable set in 'Logic variable setting'.


Register a constant value which needs to be changed later as a logic variable, and the change can be applied

to a logic by simply changing the value in the operation screen of 'Logic variable setting'.



Sctn Specifies the level of sections to output. Reads the numeric logic variable set in the operation screen of

'Logic variable setting', and outputs a numeric value (statics

type) at the specified section level.


Logic variable to read

as statistic

Select a logic variable read as statistics.


21

■ Text variable connectionGenerates fixed text or connects existing text variables to generate a connection variable.


Sctn Specifies the level of sections to output. • Outputs a text variable or input text as a text variable.

• Connects multiple texts and generates a single text variable.

• A text variable which was output by a primitive of the

previous step is overwritten with another text.

Text variable Select a text variable.

Input generated

character or input

separator for

combining charactor.

Enter a text.

If a text variable is not selected in "Text variable", a text entered for

this item is output.

Combining variable 1

to 5

Select up to five variables to connect with the first variable.

Enter a separator for "Input generated character or input separator

for combining charactor."

Extraction time Specify the time to fetch the text value by an event.

When not specifying an event, specify the section start or end time.

Last event time or

section end time

when no event.

Specify which event should be used.


start and end of a section or the event time.



Variable Overwrites a text variable output from the previous primitive step

with the specified text.


■ Numeric variable extractionExtracts the value at the time specified by an event as a step variable from a step variable, trace variable, and statistics.


Sctn Specifies the level of sections to input and output. • Extracts the step variable at the time offset from the start of a

section or subsection start time.

• When the number of seconds exceeding the section length is

specified, the section or subsection end time is extracted as

a step variable.

• When an event is specified, the value of the time when the

first event is found is extracted as step variable.

• When multiple events exist, the second and later events are

ignored.

Variable that extracts

value

Select a step variable, trace variable, and statistics.

Event at extraction

time

Specify the time of an offset from the start of a section by using a

value (-2147483648.0000 to 2147483647.0000), or select an event.

To specify the offset using a value, enter a numeric value and unit

("%" or "sec").


21

■ Execution conversionOutputs a value converted using the execution conversion table or edit conversion table registered in 'Logic variable setting'

based on a value obtained during logic execution.



Specify execution

conversion table of

logic variable

Select the execution or edit conversion table defined in 'Logic

variable setting'.

• For the execution conversion table, input variables for each

row are verified with the input condition and the data of the

cell in the specified column header of the row satisfied the

condition first is output.

• For the edit conversion table, text corresponding to the

matched ID is output.

Specify row of output

data.

Effective when the execution conversion table is selected in

"Specify execution conversion table of logic variable".

Select the header name of a row to which the execution conversion

table is output.

Input variable to

conversion table

For the execution conversion table, select a variable to enter a

value specified for each row.

For the edit conversion table, select a variable to enter a value

specified in ID column of the table.


■ Step variable filterCompare step variables or step variable and a numerical value and convert the value of a step variable value of the time

matched into the specified value.

When extracting a subsection depending on the change of a step variable, it cannot be extracted unless the

value changes in ascending order. This primitive converts the values in a range where step variable values

decrease into the same value as the one immediately before or a fixed value, and changes the order in

ascending.


Step variable of

conversion object

Select a step variable to convert. A step variable and a filter value is compared and the matched

values are converted according to the contents set in

"Conversion method".Filter value Converts the portion that matches the filter value.

If the filter value is a subsection, performs filtering process on that

subsection.

Enter a fixed value (-2147483648.0000 to 2147483647.0000) of

select a step variable.

Convert when Val. is

the same b/w before

and behind Conv.

section

If checked, conversion is performed only when the value remains

the same before and after the filter section. (For smoothing

unevenness.)

Conversion method Select one of the following options:

• Replace it by a specified value.: Replace the values of a part to

be converted by filtering with a value specified for "Replaced

specified value".

• Leave value (Exclude section head) to be converted.: Replace

the values with the value just before a part to be converted by

filtering.

When a conversion is started from the start point of a section,

values are not replaced.

Replaced specified

value

Shown when "Replace it by a specified value." is selected in

"Conversion method".

Replace the values of a part to be converted with a value (-

2147483648.0000 to 2147483647.0000).


21

■ Waveform conversion

• Setting items when selecting "MA" for "Conversion type"

Outputs the moving average trace variable.


Sctn Specifies the level of sections to output.

Conversion type Select one of the following options: "MA", "Waveform

differentiation", "Waveform integration", "Waveform difference",

"Auto-scaling", "Range-scaling", "Centering", "Wave statistic

operation", or "Logarithmic conversion".


Average point Enter the number of moving averages by using an odd number (3 to

2147483647).

A smoothed waveform (trace variable) is output.

a Select a trace variable.

Exclude maximum

value and minimum

value

Select this to output the moving average excluding the maximum

and minimum values.

Calculation option Select "Polynomial suitable", "Median", or "MAD".

• Polynomial suitable: Performs smoothing using values fit by a

polynomial of the order specified in "P-order".

• MAD: Calculates the moving average after filtering values the

using the median absolute deviation filter and removing the

outliers.

P-order Shown when "Polynomial suitable" is selected for "Calculation

option".

Select an order.

'0', '2', '4', '6', and '8' can be specified for the order.

Order = 0 is equivalent to the simple moving average.


• Setting items when selecting "Waveform differentiation" for "Conversion type"

Outputs the differential waveform of the trace variable.

n-degree differential with the specified degree "n" is performed.

• Primary differential calculates the slope between two adjacent points.

• Secondary differential further performs primary differential on the result of the primary differential.

• Tertiary differential further performs primary differential on the waveform obtained from the secondary

differential.

• n-order differential further performs primary differential on the waveform obtained from the "n-1"-order

differential.

• Setting items when selecting "Waveform integration" for "Conversion type"

Outputs a waveform obtained from time integral of trace variable data (i.e., area enclosed by the horizontal axis and

waveform).

A negative integral value is obtained where the vertical axis is formed within the negative region.

• Setting items when selecting "Waveform difference" for "Conversion type"

Extracts chronological changes of trace variable.


Degree Enter the degree of differentiation (1 to 5). A differential waveform (trace variable) is output.



a Select a trace variable. An integral waveform (trace variable) is output.


Difference point Enter the number (0 to 2147483647) of points to shift a waveform in

X axis direction in order to calculate a difference.

• The difference between a value in Y axis of a waveform

shifted for the number of points set for "Difference point" and

a value in Y axis of a waveform not shifted is output as a

trace variable.

• The original waveform is displayed for the differential number

of points from the lead of the section.



21

• Setting items when selecting "Auto-scaling" is selected in "Conversion type"

Standardizes the Y value of trace variable.

With the average of "n" Y-axis values before conversion included in the conversion target section as "Yave"

and variance as " Y", the converted value Yi' at the "i"th measurement point Yi can be expressed by the

following equation:

Yi' = (Yi-Yave)/ Y

• Setting items when selecting "Range-scaling" for "Conversion type"

Normalizes the Y value of trace variable.

With the maximum value of "n" Y-axis values before conversion included in the conversion target section as

"Ymax" and the minimum value as "Ymin", the converted value Yi' at the "i"th measurement point Yi can be

expressed by the following equation:

Yi' = (Yi-Ymin)/{(Ymax-Ymin)/2}-1

• Setting items when selecting "Centering" for "Conversion type"

Converts data into a waveform where the average value of Y values of trace variable data is '0'.

With the average of "n" Y-axis values before conversion included in the conversion target section as "Yave",

the converted value Yi' at the "i"th measurement point Yi can be expressed by the following equation:

Yi'=Yi-Yave


a Select a trace variable. The Y value of a section waveform is output as a trace variable

by converting the value 'Average value = 0' and 'Variance = 1'.


a Select a trace variable. The Y value of a section waveform is output as a trace variable

by converting it into a waveform of which the maximum value is

'1' and the minimum value is '-1'.


a Select a trace variable. The Y value of a section waveform is converted into a

waveform (trace variable) on which the average is '0'.


• Setting items when selecting "Wave statistic operation" for "Conversion type"

Perform calculation between trace variable and trace variable, or trace variable and statistics.

Precautions

The following precautions apply when "Wave statistic operation" is selected for "Conversion type":

• Variables specified in "a" to "h" must be used in 'Expression'. If not used, an expression error occurs.

• The function 'rand()' that can be used in 'Expression' must take an argument (e.g., rand(a)). Otherwise, a syntax error

occurs. Note that the specified argument is not used within the function.

• Setting items when selecting "Logarithmic conversion" for "Conversion type"

Performs logarithmic conversion on trace variable.


a to h Select one or more trace variables. And then select trace variable,

statistics, or step variable.

A waveform (trace variable) of the operation result is output. If

there is an operation which cannot be performed, '0' will be

output as the result.[Syntax] button Shows explanation of the expression syntax in a pop-up window.

(1) Expression Enter exponentiation, multiplication, division, addition, subtraction,

and functions to describe an operation expression (up to 256

characters).

When describing an expression, use trace variables specified by "a"

to "h", statistics, and step variables.

Refer to the pop-up window displayed by clicking the [Syntax]

button for the available functions.

Expression error It is lighted in red if the entered operation expression is incorrect.


(1) Logarithm type Select "Common", "Natural", or "Base=2". A waveform which is converted in a logarithmic conversion for

"Common", "Natural", or "Base =2" is output.a Select a trace variable.

(1)

(1)


21

■ Waveform interpolationInterpolates the sampling interval of trace variable to the minimum of 10 msec.

When analyzing trace variables of which the sampling intervals are different by overlapping them, they are is

converted into the waveform with a finer granularity.


Select supplemented

variable

Select a trace variable. • A selected trace variable is converted into a waveform with a

specified sample interval, then it is output as a trace variable.

• Assign the most recent original sampling value to the sample

to generate for interpolation.msec Enter a sampling interval (10.0000 to 2147483647.0000 after

conversion.

Enter a value smaller than the original interval before conversion.

(10 ms or longer)


■ Shift in X axisShift the X (time) axis of trace variable forward or backward.


Sctn Specifies the level of sections to output. A variable is output as a trace variable by shifting X (time) axis

of the data of the trace variable forward or backward.Variable shifted to

time axis direction

Select a trace variable.

Shift point Enter the number (-2147483647 to 2147483647) of points to shift a

variable in X axis direction.

A waveform is shifted right by entering a positive number and

shifted left by entering a negative value.


21

■ Max. Avg. Min. waveform ext.Extracts a waveform connecting the average, maximum, and minimum values of each point from waveforms of multiple trace

variables.


Output form Select one or more types of waveform to output from "Ave", "Max",

and "Min".

• Ave: Average value

• Max: Maximum value

• Min: Minimum value

A waveform (trace variable) connecting the average values,

maximum values, or minimum values of each point is output.

The output variable name is as follows:

• Ave: Output variable name_Average

• Max: Output variable name_Max

• Min: Output variable name_MinTrace data Specify multiple trace variables to extract waveforms.


■ Linear curve fitPerforms a linear fit and extracts curve fit parameters.


Fitting section Specify two sections to perform a liner fit. A linear fit (least-squares method) is performed and a fit

waveform (trace variable), slope (statistics), intercept

(statistics), and MSE (Mean Square Error) (statistics) are

output.

Curve fitting variable Select a trace variable.


21

■ Polynomial curve fitPerforms a polynomial fit and extracts curve fit parameters.


Fitting section Specify two sections to perform a polynomial curve fit. A polynomial fit is performed and a fit waveform (trace

waveform), coefficient (statistics), and MSE (Mean Squared

Error) (statistics) are output. The intercept of a coefficient is 1.

Coefficients are output in an order where orders are few.

Curve fitting variable Select a trace variable.

P-order Specify the order of the polynomial (0 to 2147483647).


■ Exponential curve fitPerform an exponential fit and extracts curve fit parameters.


Fitting section Specify two sections to perform exponential curve-fitting. An index fit is performed and a fit waveform (trace variable),

amplitude (statistics), dumping (statistics), and MSE (Mean

Squared Error) (statistics) are output.Curve fitting variable Select a trace variable.


21

■ Individual trace statisticExtract a statistic in a section or subsection from a trace variable.


Sctn Specifies the level of sections to output. Outputs the following according to option settings for the

section level output.

• Outputs at the subsection level.: Statistics of all subsections

at the selected section levels (L1, L2, and L3).

• Subsection number: Statistics of the specified subsection

number to statistics at the upper-level section level.

• Ave/Max/Min/Range/Sigma: Central statistics value (one of

the average, maximum, minimum, range, or sigma) at the

selected section levels (L1, L2, and L3) to statistics at the

upper-level section level.

Central value Select one of the following options:




• Range: Range

• Sigma: Standard deviation

• Variance: Variance

• Median: Median

• Mode: Mode

• RMS: Root mean square

• Kurtosis: Kurtosis

• Skewness: Skewness

• Int-val.: Integral value

• Interval1: Time from the minimum value to the next maximum

value (sec)

• Interval2: Time from the maximum value to the next minimum

value (sec)

• Interval3: Time between the maximum value and minimum value

(sec)

• Start value: Start value of section

• End value: End value of section

• Corr-range.: Difference between the maximum and minimum

residual values

• Intermediate: Intermediate value of section


statistic


Section level output

option

Shown when "L1", "L2", or "L3" is selected in "Sctn".

Select one of the following options: "Outputs at the subsection

level.", "Subsection number", "Ave", "Max", "Min", "Range", or

"Sigma".

Subsection No. Shown when "Subsection number" is selected in the option for the


Specify the subsection number (1 to 2147483647).


■ Event specified trace value ext.Extracts the value at the time when an event occurs from multiple trace variable data.


Sctn Specifies the level of sections to output. • The value of the occurrence time when the first event is

found in a specified section in the data of multiple trace

variables is extracted as the statistic.

• If no event is specified, the time value obtained by adding the

offset time to the start time of a specified section is output as

the statistics.

Variable name that

extracts value

Select one or more trace variables.

Extraction time event Select an event name.

If an event is not found, the value at the head of section is extracted.

Offset (sec) Specify the offset time (-2147483648.0000 to 2147483647.0000)

from the time of an event occurrence or the start time of a section.

When specifying a negative value, a value of the time before the

event occurrence is extracted. If it is before the start time of a

section as a result of an offset, the value of the start time of the

section is extracted.

A negative offset when no event is specified is reset to "0" when the

primitive is added or updated.


21

■ Multi trace statisticExtracts statistics from multiple trace variable data by specifying the extraction range.



Sctn Specifies the level of sections to output. • Extracts statistics with the name 'primitive output name' +

'trace variable name' + 'statistic' from a trace variable within

the specified range.

• The length of the section from which statistics were extracted

is output as statistics.

Specify and extract

event or text

• When specifying the range by events, select the start and end

events.

• When specifying the range by text and step variables, specify a

variable name, comparison operator, and comparison value (up to

32 characters).

If multiple ranges exist, statistics is output for the first range. If the

specified range is not found, nothing is output.

If this item is not specified, statistics for the start and end of the

section is output.

Section level output

option

Shown when a subsection ("L1", "L2", or "L3") is selected in "Sctn".

Select one of the following options: "Outputs at the subsection

level.", "Subsection number", "Ave", "Max", "Min", "Range",

"Sigma", or "All up to specified sub-sec No.".

Outputs the following according to the setting.

• Outputs at the subsection level.: Outputs statistics of all

subsections at the selected levels (L1, L2, and L3) at the

subsection level.

• Subsection number: Outputs statistics of the specified subsection

number only as the statistics at the section level.

• Ave/Max/Min/Range/Sigma: Outputs the central value (one of the

average, maximum, minimum, range, or sigma) of the statistics at

the selected levels (L1, L2, and L3) as statistics at the section

level.

• All up to specified sub-sec No.: Outputs all statistics of

subsections from the first subsection to the specified subsection

at the section level.

Subsection No. Shown when "Subsection number" is selected in the option for the


Specify a subsection number (1 to 10000).

Central value Select one or more central values to extract from the following:




• Range: Range



• Median: Median

• Mode: Mode




• Int-val.: Integral value

• Interval1: Time from the minimum value to the next maximum

value (sec)

• Interval2: Time from the maximum value to the next minimum

value (sec)

• Interval3: Time between the maximum value and minimum value

(sec)

• Start value: Start value of section

• End value: End value of section

• Corr-range.: Difference between the maximum and minimum

residual values

• Intermediate: Intermediate value of section

Trace data Select one or more trace variables.


21

■ Area statisticCalculates time during which the trace variable remains within the specified Y value range.


Sctn Specifies the level of sections to output. The time of a trace variable within the specified range and the

% value within the specified section are output as statistics.Section output. Outputs area statistics extracted in the first deepest subsection (L1,

L2, or L3) at the section level.


area statistic


Specify range to

extract area statistic.

Select "Upper limit excess," "Less than Lower limit," or "Range of

limit pair."

• Upper limit excess: Exceeded the upper limit.

• Less than Lower limit: Falls below the lower limit.

• Range of limit pair: Stays within the range between the upper and

lower limits.

Upper limit Shown when "Upper limit excess" or "Range of limit pair" is selected

in "Specify range to extract area statistic."

Enter an upper limit value (-2147483648.0000 to

2147483647.0000) or select a variable.

When entering the upper limit value, it should be greater than the

lower limit (upper limit > lower limit).

Lower limit Shown when "Less than Lower limit" or "Range of limit pair" is

selected in "Specify range to extract area statistic."

Enter a lower limit value (-2147483648.0000 to 2147483647.0000)

or select a variable.

When entering the upper limit value, it should be greater than the



■ Event statisticExtracts the section length, frequency, and interval as statistics.


Sctn Specifies the level of sections to output. The following are output as statistics according to the setting in

"Central value".

• Event section: Outputs the specified section length.

• Event frequency: Extracts the number of event occurrences

within the section. Also counts events overlapping the

section start and end times.

• Event interval: Calculates the event occurrence interval by

dividing the section length by 'number of events which

occurred + 1'.

If no event exists within the section, the section length is

regarded as the event interval.

Also counts events overlapping the section start and end

times.

Central value Select "Event section", "Event frequency", or "Event interval".

Event that extracts

statistic

When "Event frequency" or "Event interval" was selected in "Central

value", select the event to extract the statistics.


21

■ Range specified event statisticExtracts multiple sections defined by start and end events and calculates the statistics of the section length.


Sctn Specifies the level of sections to output. • Extracts the start-to-end section from the specified event and

outputs the statistics for the section length.

• Include the events overlapping the section start and end

times.

• When it is successful, 'Boolean variable = 0' is output. When

it fails, 'Boolean variable = 1' is output.

[Start event name]

button

Select the start event of the section to extract the statistics.

Start offset (sec) Specify an offset (-2147483648.0000 to 2147483647.0000) from the

start event.



More than 2 start

candidates.

Select the event to use between "First" and "Last" if multiple events

satisfying the specified condition exist.

[End event name]

button

Select the end event of the section to extract the statistics.

End offset (sec) Specify an offset (-2147483648.0000 to 2147483647.0000) from the

end event.



More than 2 end

candidates.

Select the event to use between "First" and "Last" if multiple events

satisfying the specified condition exist.

Calculate central

value

Select statistics type to output from the following: (Multiple items

can be selected.)

• Average section length: Average section length

• Maximum section length: Maximum section length

• Min section length: Minimum section length

• Max section length - Min section length: Range



• Section #: Number of sections

• Total section length: Total section length


■ Event specified variable value ext.Extracts the value at the time when an event occurs from multiple step variables.


Sctn Specifies the level of sections to output. • The value of the occurrence time when the first event is

found in a specified section in the data of multiple step

variables is extracted as the statistic.

• If no event is specified, the time value obtained by adding the

offset time to the start time of a specified section is extracted

as the statistics.

State variable Select a step variable (state variable) to extract a value.

Extraction time event Specify the extraction time by an event.

If an event not found, the value of the head in a section is returned.

Offset (sec) Specify an offset from the event occurrence time (-

2147483648.0000 to 2147483647.0000).

If the event is not specified, the specified value is regarded as an

offset from the start time of the specified section. A negative offset

when no event is specified is reset to '0'.

If the offset time comes after the end of the section, the section end

time is returned.


21

■ Multi-section totalCalculates single statistics from multiple statistics for the current and past section data.


Calc. section grouping

Var.

Select this when grouping sections by the same text variable value. • Calculates the specified statistics for the specified calculation

section and outputs the result.

• When outputting, 'Boolean variable = 1' is output. When not

outputting, 'Boolean variable = 0' is output.Calc section Specify the number (1 to 10000) of sections in which a calculation is

performed.

(1) Calculation type Select one of the following options:

• [Calculate with Past Sections Only]: Uses values of the past

sections (those before the currently running section) for

calculation.

• [Include Current Section in Calculation]: Include the current

section value in addition to the past section values for calculation.

Central value Select statistics from the following: (Multiple items can be selected.)




• Range: Range



• Median: Median

• Mode: Mode




• Additional value: Additional value (integral value)

Section statistic Select from calculation target from the statistics variables and

execution result (boolean variables) extracted by the logic in the

previous primitive step.

Reset by 0 values. If checked, when statistics has a value of '0' in the current section,

the calculation result is reset to '0' at that section, and calculation

resumes with the next section.

Output from that section is '0'.

(1)


■ Subordinate section totalSummarizes the current subsection statistics and outputs the result as the statistics of the upper-level section.


Output level of result Specify the level of sections to output the calculation result. • Calculates the subsection statistics at the lower level and

outputs the result as the statistics of the upper-level section.

• When "Additional value(sum)" is selected in "Central value",

all the statistics of the lower level sections are aggregated

and the result is output as the statistics of the upper-level

section.

Variable that

calculates statistic of

subsection

Select a statistic in a level lower than a section to be output or a

boolean variable.

Central value Select the central values to aggregate from the following:




• Range: Range



• Median: Median

• Mode: Mode

• Number of effective data: Number of subsections with values

calculated

• Additional value(sum): Statistics or total of boolean variables


21

■ Multi-section calculationPerforms an operation on statistics of the past and current sections.


Calc. section grouping

Var.

Select this when grouping sections by the same text variable value. • Performs an arbitrary operation between the statistics of the

'n'th last section and that of the current section.

• When there is past data in a section which was specified for

all variables, 'Boolean variable = 0' is output.

• When there is no past section data which was specified for

one or more variables, 'Boolean variable = 1' is output.

• Past sections without statistics are skipped.

• If the output from this primitive itself does not exist, '0' is used

instead.

• If the past statistics does not exist for any of "a" to "g", the

operation result is output as null.

• If there is an operation which cannot be performed in an

operation expression, 'null' will be output as the result.

a to g Selects a statistics name.

By selecting "Forward of n section", output from this primitive for the

'n'th last section can be referenced.

(1) Reading section

(Input fields for 'a' to

'g')

Enter a number (0 to 100) for a reading section.

'0' refers to the current section and a number 'n' refers to the 'n'th

last section.

[Syntax] button Shows the expression syntax.

Expression Enter exponentiation, multiplication, division, addition, subtraction,


characters).

When describing an expression, use statistics specified by "a" to

"g".




(1)


■ Calculation in present sectionPerform an operation between statistics in the current section.


Sctn Specifies the level of sections to input and output. • The calculation result acquired from a statistic according to

the specified output type is output as a statistic.

• If there is an operation which cannot be performed in an

operation expression, 'null' will be output as the result.

a to h Selects statistics.

Output type Select the type of statistic to be output.

• By specifying "Expression", the input field for an operation

expression is displayed. The calculation result of the operation

will be output.

• By selecting "Ave", "Max", "Min", "Range", or "Total", the selected

type of statistic is calculated from a variable selected from "a" to

"h". Then, the calculation result will be output.

[Syntax] button Shows the expression syntax.

Expression Enter exponentiation, multiplication, division, addition, subtraction,


characters).

When describing an operation expression, use statistics specified

for "a" to "h".





21

■ Raises level to section statistic.Upgrades the level subsection statistics to section statistics.


Sctn Specify the subsection of the subsection statistics which are

upgraded to section statistics.

• Outputs the specified statistics as the statistics at the section

level.

• The output statistics name is 'output variable name_statistics

name'.Level up statistic Select a statistics name at L1 or lower level.

(1) Subsection No. For the L3 statistics, specify the L1, L2, and L3 section numbers (1

to 2147483647).

For the L2 statistics, specify the L1 and L2 section numbers (1 to

2147483647).

For the L1 statistics, specify the L1 section number (1 to

2147483647).

(1)


■ Statistic level downDowngrades the level subsection statistics to a lower level.


Sctn Specifies the level of sections to input and output. Outputs the specified statistics as the statistics at a lower

section level with a different name.Level down statistic Select a statistics name to downgrade.


21

■ Regression operationRun a regression equation created by a partial matrix.


Matrix including

regression equation

Specify the file path (up to 256 characters) of a partial matrix in

which 'Recurrence calculation' was performed in 'Multiple

regression analysis (LMR)'.



The partial matrix is stored at the following location under the name

'(partial matrix name).csv':

(Matrix storage folder)\Attribute\(Whole matrix name)

A predicted value calculated based on the regression

expression is output as statistics.

Regression eq. Select a regression equation for Multiple Regression Analysis

(LMR).

(1) Regression

equation list

List of statistics names and partial regression coefficients of

regression equations is shown.

If it does not match the statistics name extracted in the primitive

step before the logic, it is grayed out. In this case, the equation is

run in an incomplete format.

(1)


■ MVA statisticPerforms MT distance calculation based on the partial matrix.


Criterion matrix Specify the file path (up to 256 characters) of a partial matrix in

which 'MT distance calculation' is performed in 'Maharanobis-

Taguchi method' and the validity of the Mahalanobis distance is

checked.



The partial matrix is stored at the following location under the name

'(partial matrix name).csv':

(Matrix storage folder)\Attribute\(Whole matrix name)

• Mahalanobis distance, Q statistic (SPE), and Euclidean

distance (EU distance are output as statistics.

• EU distance is the Euclid distance of the diagnosis section

for the criterion matrix, and used when there is no correlation

between the variables.

Group Text variables in the matrix can be selected.

Specify it when calculating the distance by filtering the criterion

matrix to the same text value as that of the diagnosis section.

Processing type Select "MT distance".

Variable name Shows the statistics included in the partial matrix selected in

"Criterion matrix".

Select two or more statistics used for diagnosis.

Statistics not extracted from the preceding primitive steps are not

shown.


21

The following gives an example of using the primitive 'MVA statistic':

Create a logic and add primitives 'Multi trace statistic' and 'Save CSV data' to that logic.

Page 305 Multi trace statistic, Page 340 Save CSV data

Use the primitive 'Multi trace statistic' to select a variable to apply to diagnosis by 'Mahalanobis-Taguchi

method'.

Calculate the statistics of the variable selected in and save the result in a CSV file.

Create a partial matrix from the CSV file created in in 'Create correlation matrix'.


To avoid multicollinearity, perform the following operations to create a criterion matrix which is an

appropriate Mahalanobis distance.

• Delete one of high correlation coefficient variables (statistics) in combination.

• Manipulate variables (statistics) with missing or abnormal values appropriately.

Perform 'MT distance calculation' using the 'Mahalanobis-Taguchi method' to check the validity.


Repeat until a criterion matrix with a valid Mahalanobis distance is created.

Delete the primitive 'Save CSV data' from the logic defined in and define and add the primitive 'MVA

statistic' to the logic.

In the definition of the primitive 'MVA statistic', specify the partial matrix created in in "Criterion matrix".

Following the primitive 'MVA statistic' in , define the primitive 'Limit pair diagnostics' or 'SPC diagnostics'

and diagnose the output variable of the primitive 'MVA statistic'.

Page 324 Variable limit pair GB, Page 330 SPC diagnostics


■ Enhanced trace GBDiagnoses the waveform using GB (guard band).

The enhanced trace GB can be created by entering upper and lower limits or from a reference waveform.

• Creating guard band by entering upper and lower limits


Diagnostics section Specify "Sctn" or "L1".

For L1, up to ten sections can be diagnosed.

• Outputs an event at the time when the upper or lower limit

GB was exceeded at the specified section level.

• For the GB error, 'Boolean variable = 1' is output. For the

normal state, 'Boolean variable = 0' is output.

• The following items are also output.

Boolean variable ('1' for a section and '0' for L1 to L3

sections)

Trace variable (upper and lower limit values)

Pop up window If checked, opens a pop-up window upon detection of guard band

deviation.

A response data file can be output by displaying the pop-up window.

(1) n of m (number of

sections for

consecutive

deviation)

Specify a section (1 to 1000) among the number of specified

sections. It is determined as an error when the guard band is

exceeded in the specified section.

When specifying '3' for "n of" and '5' for "m", an error occurs if the

guard band is deviated in the third section within five sections.

Diagnostics variable Select a trace variable to diagnose.

Reference variable Not selected.

Upper limit excess

points

Specify the number of point (1 to 100) to define the timing of an

error determination. It is determined as an error when the upper limit

of the guard band is exceeded continuously by the specified number

of point in a single section.

Lower limit excess

points


error determination. It is determined as an error when the lower limit



Upper limit Specify the upper limit value (-2147483648.0000 to

2147483647.0000) of a guard band.

When the diagnosis section is L1, up to 10 sections can be specified

for each section number of the L1 section.

When specifying '0', GB diagnosis for the upper limit is not

performed. When specifying a value other than '0', the upper limit

should be greater than the lower limit (upper limit > lower limit).

Lower limit Specify the lower limit value (-2147483648.0000 to

2147483647.0000) of a guard band.

When the diagnosis section is L1, up to 10 sections can be specified

for each section number of the L1 section.

When specifying '0', GB diagnosis for the lower limit is not

performed. When specifying a value other than '0', the upper limit

should be greater than the lower limit (upper limit > lower limit).

Comment Specify a text (up to 100 characters) shown in the comment column

of the pop-up window.

(1)


21

• Creating guard band from reference waveform


Diagnostics section Specify "Sctn" or "L1".

For L1, up to ten sections can be diagnosed.

• Outputs an event at the time when the upper or lower limit

GB was exceeded at the specified section level.

• For the GB error, 'Boolean variable = 1' is output. For the

normal state, 'Boolean variable = 0' is output.


Boolean variable ('1' for a section and '0' for L1 to L3

sections)

Trace variable (upper and lower limit waveforms)

Pop up window Opens a pop-up window upon detection of guard band deviation.


(1) n of m (number of

sections for

consecutive

deviation)

Specify a section (1 to 1000) among the number of specified

sections. It is determined as an error when the guard band is

exceeded in the specified section.

When specifying '3' for "n of" and '5' for "m", a GB error occurs if a

deviation occurs in the third section within five sections.


Reference variable Select a trace variable used as the standard for the guard band.

Upper limit excess

points


error determination. It is determined as an error when the upper limit



Lower limit excess

points


error determination. It is determined as an error when the lower limit



GB-slope Specify the slope (Vertical axis (Y value change) Horizontal value

(Number of Points)) applied when the upper-limit guard band

descends or the lower-limit guard band ascends.

When '0' is specified, this setting will be disabled.

Unit Specify the upper/lower-limit width by "Value" or "%" of the

waveform amplitude.

Set offset. Check it to specify the pre- and post-offsets.

Pre-offset Enter a backward offset for the waveform of a standard variable as

the number of plots (1 to 2147483647).

Post-offset Enter a forward offset for the waveform of a standard variable as the

number of plots (1 to 2147483647).

Upper limit Enter the upward width (value or percentage of amplitude) for the

waveform of the standard variable. When the diagnosis section is

L1, up to 10 sections can be specified for each section number of

the L1 section.

The default value is '0'. When specifying '0', GB diagnosis for the

lower limit is not performed.

Lower limit Enter the downward width (value or percentage of amplitude) for the

waveform of the standard variable. When the diagnosis section is

L1, up to 10 sections can be specified for each section number of

the L1 section.

The default value is '0'. When specifying '0', GB diagnosis for the

lower limit is not performed.

Comment Specify a text (up to 100 characters) shown in the comment column

of the pop-up window.

(1)


■ Stability deviating GBDetect the point to which a value is input in the set range and the point which a deviation occurs.

After executing this primitive, response data file can be output by executing the primitive 'Write log and error notification'.


21


Sctn Specifies the level of sections to input and output. • In the specified section level, if trace variable values

consecutively for the number of points within the continuous

range are included in a diagnosis amplitude from the time

when the stabilization is detected, the start point is output as

an stabile point.

• When "Floating" is set for "Diag-method.", set the upper and

lower limit range by referencing the point when the stable

event is detected as the starting point. When it is deviated in

the upper and lower limit range consecutively for the number

of times set in "Number of deviating detection points", the

start point is output as a deviation event.

• When deviations occur between two points, the time of the

deviation time will be the time of the former point.

• Time from the first stabilization detection start time to the

stabilization point as well as time from the stabilization point

to the deviation point are output.

Time is not output for the second and later stabilization

points.

When the stabilization point is detected, the upper/lower-limit

GB value is output.

The above process is repeated for the specified repeat

count.

• A boolean variable (One or more is detected: 0, Nothing is

detected: 1), statistics (last deviation value, or last value of a

section if nothing is detected), event (all detected time for

stabilization and deviation time) are output.


Statistics (Number of stability, number of deviations,

diagnosis upper value, diagnosis lower value, time for

stabilization, stability time)

Variable that detects

stability and deviation


Diag-method. Select one of the following options:

• Floating: Set the diagnosis width by centering the value of the

start point of stabilization (that is a state in which values are in a

specified range) to perform a diagnosis.

• Targeted value: Set the diagnosis width by centering the standard

value to perform a diagnosis.

• Lower limit: Set the lower limit of the diagnosis width as the

standard value to perform a diagnosis.

Range of floating When "Floating" is set for "Diag-method.", select "No cliterian",

"Over base", or "Below base".

Criterian value When "Floating" is set for "Diag-method." set the threshold of an

area to detect a stable deviation according to the contents set in

"Range of floating" as follows:

• Below base: Set the upper limit value (-2147483648.0000 to

2147483647.0000).

• Over base: Set the lower limit value (-2147483648.0000 to

2147483647.0000).

When setting "No criterion" for "Range of floating", the value set for

this item will not be used.

For "Targeted value" or "Lower limit", set a value to be referenced.

Sets the standard value from statistics, logic variable, or constant.

When a step variable is set as the standard value, the value of the

first step variable at the section level is used.

Diagnostics width Specify the width between the upper limit value and lower limit value

to be diagnosed (0.0000 to 2147483647.0000).

Set a statistic, a logic variable, and a constant to a diagnosis width.

When a step variable is set as the diagnosis width, the value of the

first step variable at the section level is used.

Number of

stabilization detection

points.

Specify the number (1 to 2147483647) to detect the stabilization (a

state in which values are in a specified range). It is detected when

values are stored continuously by the specified number of point

stored in the diagnosis width.

Number of deviating

detection points.

Specify a number to detect the deviation (1 to 2147483647). When

a value deviates from the diagnosis width for the specified value,

the deviation is detected.

Frequency in which

stability and deviating

are repeatedly

detected

Specify the number of times (1 to 2147483647) to detect

stabilization within one section.

Opposite scanning Check this to perform scanning in the backward direction.

Specify detection start

time

Specifies the detection start event.


■ Variable limit pair GBPerforms a limit pair value diagnostics of the trace variable using a variable.



Sctn Specifies the level of sections to input and output. • Whether a waveform in a specified section deviates the

upper or lower limit value is checked.

• A boolean variable (Within a range: 0, Outside of a range: 1),

statistics (a value when a deviation is detected outside of a

range or last value when it is detected within a range), event

(time when a deviation is detected) are output.


Statistics (UCL and LCL)

Trace variable (UCL and LCL)


UL variable Enter a fixed integer value (-2147483648.0000 to

2147483647.0000), or specify a statistic or a logic variable (integer).

When entering a fixed integer value, the upper limit variable should

be greater than the lower limit variable (upper limit variable > lower

limit variable).

Fault when exceeding

specified continuous

points

Specify the number of point (1 to 2147483647) to define the timing

of an error determination. It is determined as an error when the

upper limit value is exceeded continuously by the specified number

of point.

LL variable Enter a fixed integer value (-2147483648.0000 to

2147483647.0000), or specify a statistic or a logic variable (integer).

When entering a fixed integer value, the upper limit variable should

be greater than the lower limit variable (upper limit variable > lower

limit variable).

Fault when falling

below specified

continuous points

Specify the number of point (1 to 2147483647) to define the timing

of an error determination. It is determined as an error when the

lower limit value is exceeded continuously by the specified number

of point.

Subsection No. of

diagnostics object

Enter the number (0 to 2147483647) of the L1 subsection to be

diagnosed.

Diagnoses the L1 subsection only even when L2 and L3 sections

exist.

When '0' is specified, all subsections are diagnosed.


21

■ Amplitude change detectionDetect an event at the change point of the waveform amplitude and a trace waveform of the amplitude change rate by using

the movement window (one window in the first half and the last half).


Sctn Specifies the level of sections to extract. • An event at the change point of the amplitude and a trace

waveform of the amplitude change rate are output.

• When the checkbox of "Save" is selected, output can be

referenced in 'Data display after logic execution'.


• The following item is also output.

Statistics (Number of detection)

Diagnostics waveform Specify a trace variable to extract an amplitude change.

Movement window Specify the width of the former half and the latter half of a point

which is focused as the change point of amplitude by using a

number of points (5 to 2147483647).

Width is the same for the former and latter parts.

N% Specify the average amplitude (0.0000 to 2147483647.0000) of the

latter window when a percentage of the average amplitude for the

former window is 100%.

Specify a value less than or more than '100'. ('100' cannot be

specified.)

Upper limit threshold

Y value

Specify the upper limit (-2147483648.0000 to 2147483647.0000) of

a threshold (Y value) so as not to detect the amplitude change of a

waveform.

An amplitude change exceeding a specified value are not detected.

When entering a threshold, the upper limit threshold Y value should

be greater than the lower limit threshold Y value (upper limit

threshold Y value > lower limit threshold Y value).

Lower limit threshold

Y value

Specify the lower limit (-2147483648.0000 to 2147483647.0000) of

a threshold (Y value) so as not to detect the amplitude change of a

waveform.

An amplitude change falling below a specified value are not

detected.




Repeat detection

times

Specify the number of times (1 to 2147483647) to detect an

amplitude change within one section.

Detection start event Specifies the detection start event.


• Operation flow to detect change in amplitude

1. A linear fit is performed to make a waveform horizontal.

2. By regarding and as a former window and a latter half window respectively and setting the average amplitude of

the former window is 100%, the average amplitude of the latter window is calculated and the amplitude change is

determined.

3. In succession, by regarding and as a former window and a latter window respectively, and change in a slope is

determined in a same way.

(1) Average amplitude

(2) Movement window interval

(2)

(1)

(1)

(1)

(1)

� � � �

(2) (2) (2)


21

■ Slope change detectionDetect an event at the change point of the waveform slope and a trace waveform of the slope change by using the movement

window (two windows in the first half and the last half).

Precautions

A waveform including noise should be smoothed by moving average in the step of the previous primitive. In addition, an

appropriate width should be specified for the movement window to capture the slope change of a waveform.


Sctn Specifies the level of sections to extract. • An event at the change point of the waveform slope and a

trace waveform of the slope change are output.

• When the checkbox of "Save" is selected, output can be

referenced in 'Data display after execution'.


• When "Slope change Max." is specified, the difference trace

waveform of the slope change is output.

• When "+- peak" is specified, the trace waveform of the slope

change is output.

• The following item is also output.

Statistics (number of detection)

Diagnostics waveform Specify a trace variable to extract an amplitude change.

Movement window

interval

Specify the width of the former half and the latter half of a point

which is focused as the change point of slope by using a number of

points (5 to 2147483647).

Width is the same for the former and latter parts.

Specify [Slope

change maximum] or

[Peak].

Select "Slope change Max." or "+- peak".

Dir. of slope change Specifies increase or decrease.

'+ peak' indicates decrease and '- peak' indicates increase.

Maximum noise

amplitude

Specify the Y value (0.0000 to 2147483647.0000) of a waveform

that is regarded as a noise and be ignored.

Upper limit threshold

Y value

Specify the upper limit (-2147483648.0000 to 2147483647.0000) of

a threshold (Y value) so as not to detect the slope change of a

waveform.

Values exceeding the threshold are not detected.




Lower limit threshold

Y value

Specify the lower limit (-2147483648.0000 to 2147483647.0000) of

a threshold (Y value) so as not to detect the slope change of a

waveform.

Values falling below the threshold are not detected.




Repeat detection

times

Specify the number of times (1 to 2147483647) to detect a slope

change within one section.

Detection start event Specifies the detection start event.


• Operation flow to detect change in a slope

When change in a slope is maximum

1. A linear fit is performed for each former window and latter window.

2. By regarding and as a former window and a latter half window respectively, change in a slope of a line to which a

linear fit is performed is determined.

3. The point where the slope decreases or increases is defined as the detection point. (When the slope does not exceed

'Maximum noise amplitude', it is regarded as noise and the change in slope is not detected.)




(1) Change detection point

(2) Maximum noise amplitude


(1)

� �

(1)

(3)

(2)

(1)

(2)

(2)

(3) (3) (3)

� � � �


21

For +- peak

1. A linear fit is performed for each former window and latter window.

2. By regarding and as a former window and a latter half window respectively, change in a slope of a line to which a

linear fit is performed is determined.

3. The point where the slope goes up from the bottom right to top right is defined as the detection point of the minus peak.

And, the point where the slope goes down from the top right to bottom right is defined as the detection point of the plus

peak. (When both the slopes of lines in a former and latter windows do not exceed 'Maximum noise amplitude', it is

regarded as noise and the change in slope is not detected.)




(1) Change detection point

(2) Maximum noise amplitude


(1)

� �

(1)

(3)

(2)

(1)

(2)

(2) (2)

(3)(3)(3)

� � � �


■ SPC diagnosticsPerforms SPC diagnosis by specifying the SPC rule.


Diagnostics variable Select statistics to diagnose. • If the statistics to diagnose has a less number of samples

than specified, the SPC diagnosis is not performed.

• A detection is started from the top of the list in "SPC Rule"

and ended at the timing when an error is detected.

• The output variable name_value, upper-limit value (UCL),

lower-limit value (LCL), target value (Target), ( ),

judgment result boolean value (SPC judgment normal = 0,

error = 1), execution result boolean value (executed = 0, not

executed = 1), judgment rule text variable (detected SPC rule

name) are output.

Group Select this when grouping sections by a text variable value.

Upper limit Select a statistic that indicates the upper management limit.

Lower limit Select a statistic that indicates the lower management limit.

Target value Select a statistic that indicates the center line.

Sigma Select a statistic that indicates a sigma.

[Select] button Select, delete, or update an SPC rule.

The same rule can be set for multiple times.

The selected list is shown in the "SPC Rule" list.

To perform a deletion or update, select a row in the "SPC Rule" list

and select "Delete" or "Update" from the pull-down list.

(1) Number of

diagnostics sections

Specify the diagnostics points for the SPC rule as 'n of m'.

Specify a section (1 to 20) among the number of specified sections.

It is determined as an error when the 'SPC Rule' is satisfied in the

specified section.

When specifying '3' to the left side of input area and '5' to the right

side of input area, an error occurs if the SPC Rule is satisfied in the

third section within five sections.

Note that a section (n) should be equal or less than the number of

specified sections (m) (Section (n) Number of specified sections

(m))

EWMA Check this to run the SPC diagnosis on the exponential weighted

moving average.

Weighting factor Sets a weighting factor (0.1000 to 1.0000).

Enter a factor to multiply the current section statistics.

(2) List • "SPC Rule" shows the name of the SPC rule selected by the

[Select] button.

• "n of" shows 'n' for 'number of diagnostics sections'.

• "m" shows 'm' for 'number of diagnostics sections'.

Preprocessing Select "None" or "Common logarithm conversion".

• None: No preprocessing

• Common logarithm conversion: Performs a common logarithm

conversion to the statistics in the current section and an SPC

diagnosis.

Pop up window When checked, shows the SPC screen as a pop-up window upon

SPC error (SPC judgment boolean =1).

Page 189 SPC


Mail sending level Do not use this.

(1)(2)


21

■ Limit pair diagnosticsDiagnoses the upper/lower-limit value of statistics.


Two methods are available for the limit pair diagnostics; 'diagnosing multiple statistics with same upper and lower limits' and

'diagnosing multiple statistics with same upper and lower limits set by logic variable'.

• Diagnosing multiple statistics with same upper and lower limits


Sctn Specifies the level of sections to input and output. • It is diagnosed whether a value of statistics to be diagnosed

deviates from the upper and lower limit values in a specified

section.

• A boolean variable is output for each diagnostics variable

(Normal = 0, Not normal = 1).

• One boolean variable is output with OR condition between all

selected diagnostics variables (Normal = 0, Not normal = 1).

Conversion table Not specified.

Diagnostics variable Select the diagnosis target statistics name or names.

Upper limit Enter a numerical value (-2147483648.0000 to 2147483647.0000),

or select a variable name used as the upper limit.

When entering a numerical value, it should be greater than the


Lower limit Enter a numerical value (-2147483648.0000 to 2147483647.0000),

or select a variable name used as the lower limit.




• Diagnosing multiple statistics with upper and lower limit values set for a logic variable


Sctn Specifies the level of sections to input and output. • Reads the execution conversion table and assign the upper/

lower-limit value to the diagnostics variable.

• The limit pair diagnostics are performed for each variable at

the timing of logic execution and the information on a

diagnostics variable deviating from the upper and lower limit

values is written to a template and output.

• The following values are output: individual diagnostics

variable, whole boolean variable, '0' for normal and '1' for

abnormal (AND of individual diagnostic variable), upper limit

value, and lower limit value.

Conversion table Select the execution conversion table defined in 'Logic variable

setting'.


Select line header Select the row header of an execution conversion table (up to 256

characters).

Operation Select a statistics name of the diagnosis target corresponding to the

row header, or delete the diagnostics variable selected in the

"Diagnostics variable" list.

(1) Diagnostics

variable list

Shows the row header of the selected diagnostics variable and the

diagnostics variable name.

UL row header Select a column header corresponding to the upper limit value of

the execution conversion table.

LL row header Select a column header corresponding to the lower limit value of the

execution conversion table.

Output template Select the template to write the diagnosis result.

To select the template, 'Read & insert template' must be defined in a

step prior to this primitive in the same logic.

Page 342 Read & insert template

Comment Select a column header corresponding to the comment column of

the execution conversion table.

(1)


21

■ Moving upper/lower limit diagnosticsExtracts statistics from moving sections and performs the limit pair diagnostics with common upper/lower-limit values.



Grouping var. Select this when grouping sections by the same text variable value. • Saves the specified statistics for the specified number of

sections.

• Reads saved the past statistics and calculates the statistics

for multiple sections.

• When Reader runs when it reads the CSV file, the saved

statistics is deleted upon start of the logic.

• When the logic is executed offline, the statistics is saved for

the specified range from the beginning each time.

• If the past data for the specified section is present for all

variables (i.e., normally calculated), boolean = 0 is output. If

the past data for the specified section is missing for at least

one variable, boolean = 1 is output.

• If there is no past section, null is output.

• In a specified calculation section, whether a value of

statistics deviates from the upper and lower limit values is

checked and a boolean variable for each variable (normal =

0, abnormal = 1).

When multiple variables are selected, one boolean variable

is output with OR condition between all selected variables

(Normal = 0, Not normal = 1).

Calc section Specify the number of sections to be calculated (1 to 10000).

Calculation Item Select statistics from the following (multiple items can be selected):




• Range: Range



• Median: Median

• Mode: Mode




• Additional value: Additional value

Section statistic Specify the statistics name targeted for calculation.

Reset by 0 values. When statistics has a value of '0' in the calculation section, the

calculation result is reset to '0' at that section, and calculation

resumes with the next section.

Output from that section is '0'.

Upper limit Enter a numerical value (-2147483648.0000 to 2147483647.0000),

or select a variable name used as the upper limit.



Lower limit Enter a numerical value (-2147483648.0000 to 2147483647.0000),

or select a variable name used as the lower limit.




■ Boolean operationPerforms a boolean operation on boolean variables. Detects the boolean variable and terminates a logic.


Output result

including subsection

at section level

If checked, the boolean variables at all levels are shown in the list.

The subsection boolean variable is the result of ORing the

subsections.

An AND/OR operation is performed between boolean variables

and the boolean variable (0, 1) is output.

Specify output section

level.

Boolean variables at the level specified here are shown in the list.

The input level becomes the output level.

Specify operator of

Boolean operation

between variables.

Select "AND", "OR", or "XOR".

Performs an operation on boolean variables.

• AND: AND operation

• OR: OR operation

• XOR: XOR operation

Not (reversal

processing)

Select this to reverse boolean variable values in the output result.

Boolean variable Shows the boolean variable at the level specified in "Specify output

section level.".

Logic closing option By selecting the checkbox, a logic is ended according to a boolean

variable value in the output result.

Not shown when the output level is subsection.


21

■ Boolean level changeChanges the boolean variable level.


(1) Section Specify the boolean variable subsection to upgrade. Converts the specified boolean variable to a section boolean

variable and output it.Level up Boolean

variable

Select a boolean variable name.

L1 No. Specify the L1 section number to be upgraded (1 to 2147483647).



(1)


■ Judge by textDetermine a text and output the result by using a boolean variable.


Sctn Specifies the level of sections to output. • Checks whether the text variables at the start time of the

specified section match.

• When the specified condition is satisfied, 'Boolean variable =

1' is output.

• When either one of the text variables is 'null', 'Boolean


• When both text variables are 'null', 'Boolean variable = 1' is

output.

Judgment text

variable

Select a text variable to judge.

Boolean output =1

when consist.

Select the comparison operator from the following:

• =: Equal to


Comparison text Select a variable name or enter a text (up to 256 characters).

Also, values in the selected dataset for the variable selected in

"Judgment text variable" are shown, and a value can be selected

from this list.




21

■ Judge by stepDetermine a step value and output the result by using a boolean variable.


Sctn Specifies the level of sections to output. • Judges a step variable at the section start time at the

specified section level.

• When the specified condition is satisfied, 'Boolean variable =

1' is output.

• When either one or both step variables is/are 'null', 'Boolean


Judgment step

variable

A step variable (statistics or logic variable) to compare.

Boolean output =1

when consist.

Select the comparison operator from the following:

• =: Equal to


• >: Greater than


• <: Less than


Comparison

numerical value

Select a variable (step variable, statistics, or logic variable) or enter

a value (-2147483648.0000 to 2147483647.0000).




■ IFInserted between steps of a logic to control execution of primitives.

Precautions

In this specification, reversing a boolean variable value by using an ELSE statement is only available. Note that it is different

from general control statements.

Nested control statements at the same section level (e.g., IF -> IF -> ELSE -> ELSE) are not allowed.

Since each output variable name must be unique within a logic, a syntax using control statements to change the value of the

same variable name (e.g, IF -> variable A output -> ELSE -> variable A output -> ENDIF) is not allowed.


Boolean variable

judged by IF sentence

Selects a boolean variable. • The steps of a primitive between IF and ELSE is performed

when a boolean variable is '1'.

• The steps of a primitive between ELSE and ENDIF is

performed when a boolean variable is '0'.


21

■ ELSEInserted between steps of a logic to control execution of primitives.

No setting item is available.

■ ENDIFInserted between steps of a logic to control execution of primitives.

No setting item is available.

■ CASEExecutes the Case statement and outputs a variable.

When directly entering a value in "Comparison value", "Operation", or "Output at case sentence failure", a wild

card character (* or ?) can be used. * Wild card character of '*' represents one or more arbitrary characters

and '?' represent one arbitrary character.


Variable name Select the variable name used for judgment of the Case statement. • The text value at the section start time is compared.

• If the comparison source variable matches the comparison

value, the CASE statement is satisfied.

• CASE statements in the list are checked sequentially from

the top, and the variable value of the CASE statement

satisfied first is output.

• When multiple events are compared, the CASE statement is

satisfied if any one of them matches.

• According to the setting of "Output type", trace variable,

event, numeric value, text, or statistics (L1, L2, or L3) is

output.

Comparison value Enter a value (-2147483648.0000 to 2147483647.0000) or a

character string (up to 32 characters) to compare with a variable

name.

When specifying a numeric value or text in "Variable name", the

value or text can be entered directly.

Output type Select the output variable type from the following options: "Trace",

"Event", "Text variable", "Step variable", "Statistic (section)",

"Statistic (L1)", "Statistics (L2)", or "Statistics (L3)".

Operation Select the output variable name from the variable name list

displayed according to the "Output type" setting.

Or, specify addition, insertion, or deletion of the list rows.

(1) List Shows a list of CASE comparison conditions.

• "Variable name" shows the variable name selected in "Variable

name".

• "Comparison value" shows the variable name selected in

"Comparison value" or entered value.

• "Output" shows the variable name selected in "Operation".

Output at case

sentence failure

Select a variable to output when CASE judgment is not satisfied.

(1)


■ Save CSV dataSaves the waveform, statistics data, and text data in the CSV format.


Sctn Specifies the level of sections to input and output. • Saves the waveform, statistics, and text variable data at the

specified section level in a CSV file.

• When "Logic execution unit" is selected in "File unit", the

section data selected in the dataset is saved for each of the

specified section levels.

• When a section is specified, all row data of the entire section

is output.

• When a subsection is specified, row data of the entire section

entire subsection is output.

(Example) When the number of sections is 30 and the

number of subsection is 10 each, data of a total of 300 rows

is output.

• Data from different section levels cannot be output

simultaneously.

• The output file name specified in "Specify CSV output file

name." is appended with "_YYmmdd".

Specify CSV output

file name.

Enter the file path (up to 256 characters) of a CSV file for an output.

The following macros can be used in the file name:

When a macro is added to the file name, an appropriate value is

inserted in each [[]].

• [[EQ]]: Fixed string 'UNIT'

• [[FLT]]: Unit name

• [[SEQ]]: Logic name

• [[START]]: Section start time

• [[END]]: Section end time

• [[Variable Name]]: Variable name

Do not specify a folder with the encryption attribute enabled as the

file destination.

File unit Select the period to write to a single file from the following:

• Logic execution unit: Sections which executed a single logic are

saved in a single file.

• Every day: Sections are grouped and saved in a single file for

each day.

• Every month: Sections are grouped and saved in a single file for

each month.

• Unit of section: A single file is created per section.

• Replace the Mat. in each section.: A single file is created per

section by swapping rows and columns (i.e., columns are

arranged in the row direction).

• Not add the date to the file name.: Select this when [[]] is used.

The date is not added to the file name. If the variable value within

[[]] is the same, the file is overwritten.

Writing option Select "Overwrite" or "Add" (to add new sections only).

Select "Add" when executing logics per day or month.

Operation Adds the selected variable to the list.

• By selecting "Delete selection line.", the variable selected in the

list is deleted.

• By selecting "Select statistic", "Select text", or "Select waveform",

all of the corresponding variables are added to the list.

(1) List • "Variable name" shows the variable name output to the CSV file.

• "Title" shows the variable name by default.

The string in "Title" can be changed by clicking the "Title" item

while the row is selected and entering an arbitrary text.

Number of residual

files

Specify the maximum number of files retained in the output

destination of a CSV file (0 to 2147483647).

Up to the specified number of files are retained and extra files are

deleted.

(1)


21

Precautions

Do not change the string in the "Variable name" column in the list. Otherwise, the system may not function normally.


■ Read & insert templateReads the template and inserts the specified text.

(2)

(1)


21

• The text variable contents must be defined as follows:

Arbitrary text [[Replacement character 1]]

• The replacement character enclosed in [[ ]] is converted into the specified value and then output. Characters

outside [[ ]] are output without conversion.

• Specifying [[replacement character%.3]] inserts the first three characters only.

• [[replacement character%5.3]] inserts two blank spaces and then three characters.


Sctn Specifies the level of sections to input and output. • Checks for the replacement string from the top of the

template and writes the variable value at the first matching

location.

• When the same replacement string appears multiple times,

the second and later strings are not replaced.

• The template inserted for each section specified by the

output level is output.

• If no template has been inserted, the output template is read.

(1) Output level Specifies the level of sections to output templates.

If a subsection is selected, templates are output for the number of

subsections.

Trim If checked, trims blank characters in the template.

Template Selects a text variable registered in 'Logic variable setting'.


When the menu is returned to "Select", the contents of the text

variable are shown.

File When using a file defining the message template, select the file.

Define the temperate using a text file as follows:

Do not specify a folder with the encryption attribute enabled.

Variable type Select one of the following options: "Variable name", "Text variable",

"Fixed text", "Logic variable" (variable type = text), "Equipment",

"Numeric variable", "Statistic", "Unit name", "Logic name", "Time

stamp (start)", "Time stamp (end)", "BOOL", or a logic variable (edit

conversion table).

• When selecting "Time stamp (start)" or "Time stamp (end)",

specify the time stamp form such as '%Y%m'.

%Y: Year

%m: Month

%d: Day

%H: Hour

%M: Minute

%S: Second

• When selecting "BOOL", specify characters to be inserted

separated by a comma when a boolean variable is '0' and '1'.

Substitution text Specify a substitution text on the template to write variables (up to

256 characters).

A character string enclosed in brackets '[[ ]]' on the template is a

substitution text.

The replacement string is read from the template and shown on the

menu.

After the decimal

point

Enter this value when "Statistic" is selected in "Variable type".

A value is rounded off to the specified number of digits (0 to

2147483647).

Operation Select the output variable name from the variable name list

displayed according to the "Variable type" setting.

Or, specify addition, insertion, or deletion of the list rows.

(2) List • "Substitution text" shows the character specified in "Substitution

text".

• "Type" shows the variable type selected in "Variable type".

• "Variable name" shows the variable name selected in

"Operation".

Nothing is displayed in the "Variable name" column, when

"Equipment", "Unit name", "Logic name", "Time stamp (start)", or

"Time stamp (end)" is selected for "Variable type".


■ Write log and error notificationWrites log data displayed in 'Diagnostic logic result display'.

Page 382 Displaying a Diagnostic Logic Result

By adding this primitive after a waveform diagnostic logic not having a function to display a pop-up window or a statistic

diagnostic logic as a setting to display a pop-up window, an error occurred in the logic can be detected in Monitor and the

response data file can be output.

Page 320 Enhanced trace GB

Page 330 SPC diagnostics


= 1:Output Specify a boolean variable.

If not checked, logs are always output.

A pop-up is displayed when a boolean variable is '1' and a log

which can be referenced in 'Diagnostic logic result display' is

saved.

Page 382 Displaying a Diagnostic Logic ResultSelect item Select the variable type written to the log list.

Select one of the following options: "Statistic", "Boolean variable"

(True=0, False=1), "Numeric variable", "Text variable", "Trace

variable", "Event", or an edit conversion table (logic variable).

Select content Shows a variable corresponding to each of the selected items.

For the edit conversion table, select the conversion table name in

"Select item" and ID defined in the conversion table in "Select

content".

(1) List • "Title" shows the variable item name written to the log.

• "Content" shows the variable name written to the log.

By selecting an item and single-clicking it, its contents can be

changed.

Pop up window Select this to display the pop-up window when a boolean variable is

'1'.

A waveform is displayed on a graph and a text on the list in the pop-

up window.


General warning

sound

If checked, gives the general warning sound of Windows.

Diagnostics waveform Select a waveform to diagnose.

A selected waveform is displayed in the pop-up window.

Comment Specify a text (up to 100 characters) to display on the comment

column of the pop-up window displayed in 'Diagnostic logic result

display'.

(1)


21

■ Write templateWrites the output template to a specified file.

*1 Set it so that the path length of a file to be output does not exceed 259 characters.


= 1:Output Specify a boolean variable.

Always output if not checked.

The template is written in a file per section or subsection.

Output template

variable

Select a template to save.

The template is created by using 'Read & insert template' in the

preceding step.


Output folder*1 Specify a folder path to write a template (up to 256 characters).

However, do not specify a folder with the encryption attribute

enabled.

File name*1 Enter an output file name (up to 256 characters), or select a logic

variable (text) or a template created n 'Read & insert template'.



■ Moving max minCreate a guard band while executing a logic.


Sctn Specifies the level of sections to output. • When "Maximum value waveform" is selected in "Specify

type of extraction waveform.", the waveform with the larger

value is output.

• When "Minimum value waveform" is selected in "Specify type

of extraction waveform.", the waveform with the smaller

value is output.


moving max/min

waveform

Specify a waveform to extract the moving maximum/minimum.

Specify type of

extraction waveform.

Select "Maximum value waveform" or "Minimum value waveform".

Shift rise of minimum

value wave and

descent of maximum

value wave backward.

Enter an offset to the positive X axis direction by using the number

of points (1 to 2147483647).

When the minimum value waveform is selected, the rising of the

waveform is sifted forward for the set number of points.

When the maximum value waveform is selected, the falling of the

waveform is sifted forward for the set number of points.

Shift the rise of

maximum value

waveform and the

descent of minimum

value waveform

forward.

Enter an offset to the negative X axis direction by using the number

of points (1 to 2147483647).

When the minimum value waveform is selected, the falling of the

waveform is sifted backward for the set number of points.

When the maximum value waveform is selected, the rising of the

waveform is sifted backward for the set points.

Change inclination of

rise and descent.

Enter a value to change the slope (0.0000 to 2147483647.0000).

A larger value causes a steeper slope.


21

■ Parametrical conversionConverts I/O data conversion using the conversion table.

Precautions

When the conversion table has been updated, this primitive must be updated.


Sctn Specifies the level of sections to input and output. • Performs linear transformation from an input variable to an

output variable by referencing the conversion table.

• Outputs a (statistics type) numeric value with the column

header of the conversion table as the variable name.

• Outputs '0' for an input value exceeding the range defined for

the conversion table.

Input variable to

conversion table

Specify the statistics.

Conversion table Specify a file path (up to 256 characters) of the conversion table in

the CSV format.

A conversion table is a matrix of input values and output values, and

the 1st column contains an input value and 2nd and later columns

contain output values.

The following shows the example of a conversion table.


■ Step variable generationConverts an event or value and time to a step variable.

Two methods are available to generate a step variable; converting multiple events to step variable and generating step

variable from value and time.

• Converting multiple events to step variable

Precautions

The following precautions apply when converting multiple events into a step variable:

• If a section start event is specified as the first event, input numeric values are assigned to the section start time through the

next event. If the first event is not a section start event, the numeric value of the last event of the preceding section is

retained from the section start to that event.


Specify event Not checked. Assign input numeric values to the specified event through the

next event.Event converted into

state variable

Select an event to convert to a step variable.

Up to ten events can be specified.

Numeric Enter a numeric value assigned to each event (-2147483648.0000

to 2147483647.0000).


21

• Generating step variable from value and time


Specify time (sec) Changes to "Specify Time (Sec)" if the checkbox of "Specify event"

is selected.

Creates a step variable from the specified numeric value and

time to retain the value.

No.1 to No.10 Enter time to retain each numeric value (-2147483648.0000 to

2147483647.0000 seconds).

Up to ten values can be entered.

Numeric Enter a step variable value (-2147483648.0000 to

2147483647.0000).

Up to ten values can be entered.


■ Boolean judgment statistic savingSave a value specified according to a boolean variable value as the statistics.


Sctn Specifies the level of sections to input and output. • Save a statistic specified according to a boolean variable

value.

• The saved statistics name is the output variable name of this

primitive.

Boolean variable

used to judge

Selects a boolean variable.

Value to save when

Boolean variable = 0

Select a variable name.

• When selecting "Input fixed value", enter a numeric value in

"Fixed value 1" (-2147483648.0000 to 2147483647.0000).

If both a variable name and fixed value are input, the variable is

used.

• If "Not save anything." is selected, null is saved.

• For null, nothing is shown on the chart and the variable is also

excluded from statistics calculation.

Value to save when

Boolean variable = 1

Select a variable name.

• When selecting "Input fixed value", enter a numeric value in

"Fixed value 2" (-2147483648.0000 to 2147483647.0000).

If both a variable name and fixed value are input, the variable is

used.

• If "Not save anything." is selected, null is saved.

• For null, nothing is shown on the chart and the variable is also

excluded from statistics calculation.


21

■ Read measurement dataSave a value specified according to a boolean variable value as the statistics.


Sctn Specifies the level of sections to input and output. • Compares the key item of CSV data and the text variable at

the start time of the section, and outputs the corresponding

CSV data as statistics when they match.

• If a string variable is set for "Key item 2", it is ANDed with

"Key item 1" for comparison.

Key item 1 Select a string variable to compare.

Key item 2 Select a string variable to AND with "Key item 1".

This item is optional.

Specify data folder Select the folder path in which CSV data (measurement data) is

saved (up to 256 characters).

However, do not specify a folder with the encryption attribute

enabled.

Move corresponding

measurement data to

Finish folder.

Check to create a subfolder named 'Finish' under the measurement

data folder and moves matching measurement data to that folder.

Title line Shows data for 20 rows from the first file in the selected folder.

A row corresponding to the variable name is selected from them.

One key item row Selects a column containing the key item 1.

Two key item rows Selects a column containing the key item 2.

Data Shows the variable name in the title row.

Selects all variable names to read from them.


■ Frequency filterApply the frequency filter to the trace variable waveform.

Precautions

Enter values in "Cut-off frequency 1[Low level]" and "Cut-off frequency 2[High level]" so that the values maintain the following

relationship:

• Cut-off frequency 1[Low level] < Cut-off frequency 2[High level]


Filtered variable Select a trace variable. Outputs the filtered trace variable.

Filter type Select one of the following: "Low pass", "High pass", "Band

pass", or "Band stop".

• Low pass: Low pass filter

• High pass: High pass filter

• Band pass: Band pass filter

• Band stop: Band stop filter

Cut-off frequency 1[Low level] Specify the low-pass cut-off frequency (0.0000 to 100000.0000).

Cut-off frequency 2[High level] • When "Band pass" or "Band stop" is selected for "Filter type",

specify the high-pass cut-off frequency (0.0000 to

100000.0000).

• Not required if the filter type is "Low pass" or "High pass".

(Ignored if specified.)

FIR Select this to use the FIR (Finite Impulse Response) filter.

Tap number Specify the number of taps (0 to 1000).

IIR Select this to use the IIR (Infinite Impulse Response) filter.

Topology Specify the topology.

Degree Specify a degree (0 to 100).


21

■ Power spectrumCalculates the power spectrum of the trace variable waveform.


Spectrum type Select "Auto spectrum" or "Cross spectrum".

• Auto spectrum: Auto power spectrum

• Cross spectrum: Cross spectrum

Calculates the power spectrum and outputs the result as the

spectrum waveform.

Window Enter the window information used to calculate STFT.

If "Rectangle (none)" is selected, windowing is not performed.

Trace data When "Cross spectrum" is selected in "Spectrum type", select two

variables.

Scale Select the output result unit between "Linear" or "dB".


Executing logicThere are two methods to execute a logic: 'manual execution' and 'cooperative execution with Reader and Monitor'.

The execution result of a logic is saved per unit.

■ Manual executionManually execute all primitives within the logic for the period and section of the selected dataset.

Page 384 Management

Operating procedure

1. Right-click on the primitive list and choose [Execute logic].

2. If the checkbox of [Display definition pop up.], which is on the right-click menu, is selected, the execution result screen

appears after logic execution.

A variable with 'In' shown in "Step" is an input variable of the primitive.

A variable with 'Out' shown in "Step" is an output variable of the primitive.

When more than one input or output variables have the same name, only one of them is shown on the

execution result screen. In addition, variables other than boolean variables are highlighted.


21

3. When the execution result screen appears, the [Move highlight step.] button changes to the [Run highlight step.] button.

Clicking [Run highlight step.] button to run from the primitive of which "STP" is "1" through the primitive selected when this

button is clicked.

In the following example, primitives of which "STP" is "1" and "2" are executed.

Precautions

If a period and section are not selected in the dataset, nothing happens when [Execute logic] is selected.

If the up button of the [Run highlight step.] button is clicked while no primitive is selected, no step is selected

and the logic is not executed. Clicking the down button selects the primitive of which "STP" is "1" and this

primitive is only executed.

■ Cooperative execution with Reader and MonitorWhen Reader reads a CSV file, all primitives in the logic are executed for the data (sections) in the read CSV file.

However, the following condition must be satisfied to execute the primitives:

• The created logic is shown in "logic" in the logic information list of Monitor and "State" shows "Diagnosing".


Page 416 Monitor

35621 Analyzer21.11 Setting logic variable

21.11 Setting logic variableSet variables of which values are referenced by primitives in a logic.

When a logic variable is updated, the update is immediately reflected on logic execution.

Use a logic variable when the same value is used by multiple logics or the value referenced by the logic is frequently changed

after registering the logic.

Precautions

A logic variable referenced by a logic cannot be deleted.

Operating procedure

1. Select [Analysis logic - diagnostic logic operation] [Logic variable setting].

2. Select the type of the logic variable to add and configure it.

Selecting and setting variable type

21 Analyzer21.11 Setting logic variable 357

21

Screen configurationThe following shows the screen configuration for setting a logic variable.

Displayed items

Item Description

Dataset Shows a list of dataset names.

When a dataset is selected, the equipment name associated with the selected dataset is

shown in "Equipment".

To show a dataset name, the dataset must be registered in advance.

Page 384 Management

Equipment Select the equipment name 'UNIT' to register the logic variable.

Variable type Select the type of the logic variable to add.

Page 358 Selecting and setting variable type

Numeric (fix) Numeric variable.

Set a single value for the variable name.

A logic primitive specifies a variable name and reads and uses its value.

Text (fix) Text variable.



Integer (fix) Integer variable.



Exe. Conv. table (exe.) Sets up a variable value list for a variable name, and selects a value to use from the list

when registering a primitive.

Edit Conv. table (edit) Sets up an I/O conversion table of which values are dynamically taken and used during

logic execution.

Variable name Set a variable name (up to 32 characters).

Input variable value Enter a variable value (for a numerical value: -2147483648.0000 to 2147483647.0000, for

a character string: up to 1000 characters).

(1) List of variables Shows information related to the registered logic variable.

Variable name Shows the variable name.

Variable Shows the variable value.

Variable type Shows the variable type.

Comment on each equipment A comment can be entered for the logic variable setting itself (up to 1000 characters).

Enter a comment, right-click on the variable list, and select [Save].

Comment on each variable A comment can be entered for each variable (up to 1000 characters).

Enter a comment, right-click on the variable list, and select [Add&update (highlight line)].

(1)


Selecting and setting variable typeSelect the type of the logic variable to add and configure it.

The following types of logic variables are available and the setting method differs for each of them.

Precautions

• After setting completes, right-click the variable list and select [Save]. Otherwise, the setting will not be applied.

• Restarting Monitor is required after adding or changing a logic variable.


Numeric (fix) Numeric variable.



Page 359 Setting numeric

value/text/integer

Text (fix) Text variable.



Integer (fix) Integer variable.



Exe. Conv. table (exe.) Sets up a variable value list for a variable name, and selects a value to use from the list when

registering a primitive.

Setting up

execution conversion table

Edit Conv. table (edit) Sets up an I/O conversion table of which values are dynamically taken and used during logic

execution.

Page 363 Setting up edit

conversion table


21

■ Setting numeric value/text/integer

Operating procedure

1. Select "Numeric (fix)", "Text (fix)", or "Integer (fix)" for the variable type.

Enter a character string in "Variable name" (up to 32 characters), and click the [OK] button.

2. To change a variable value, enter a character string in the "Input variable value" field (up to 1000 characters) or a value (-

2147483648.0000 to 2147483647.0000).

3. Right-click the variable list while a variable is selected, and select [Add&update (highlight line)].



■ Setting up execution conversion table

Operating procedure

1. Select "Exe. Conv. table (exe.)" for the variable type.


2. Shows the definition screen.

3. To specify a text variable value and output data from the conversion table when an input value matches this value, right-

click on "Variable value (Drug variable to input list)." and select a text variable from [Select conversion table input

variable name.].

4. When the variable value is displayed, drag it into the list on the [Input value] tab.


21

5. When using a value not shown in "Variable value (Drug variable to input list).", right-click on the list on the [Input value]

tab and select [Add on input value list.].

6. Enter an arbitrary value and click the [OK] button.

7. The entered value is shown on the list on the [Input value] tab.

8. After setting the text variable value, select "New" for "Specify line." and click the [Input completion] button.

A row is added to the conversion table.

9. Right-click on the header row of the conversion table, then select [Row header operation] [Add row item of numerical

value.] or [Add row item of character.].

10.Enter a column item name for the conversion table and click the [OK] button.

• The range of numeric values available in column items is '-2147483648.0000' to '2147483647.0000'.

• The range of character strings available in column items is 32.

11. Select a row to define a value, click the cell to define the value, and set the value directly.


12. In the "Default" row, define a value which is used when the entered value is not matched to the contents set in all rows

(list in the [Input value] tab).

While the "Default" row is selected, the [Input value] tab is not editable.

13. To delete a row, right-click the row and select [Row header operation] [Delete row.].

14. To edit setting contents of an added row (list in [Input value] tab), select the row directly on the conversion table, or enter

the row number in [Specify line.] to select the row, and edit the contents.

After editing, click the [Input completion] button.

15.After setting is complete, click the [Done] button to register the conversion table.

16. To edit the conversion table again, right-click the variable list while a variable is selected, and select [Display definition

pop up.].


21

■ Setting up edit conversion table

Operating procedure

1. Select "Edit Conv. table (edit)" for the variable type.


2. Shows the definition screen.

3. Enter an ID (up to 32 characters) and a text (up to 1000 characters), right-click on the ID list, then select [Add] or [Insert].

4. To read a CSV file for setting, right-click on the ID list and select [File] [File] to select the CSV file.

(1) ID entry area

(2) Text entry area

(3) ID list

(1)(2)

(3)


5. "ID" and "Text" are displayed in the ID list.

An added or inserted item can be deleted by selecting it in the ID list, right-clicking it, and then choosing

[Delete].

6. After setting is complete, click the [Done] button to register the conversion table.

7. To edit the conversion table again, right-click the variable list while an item is selected, and select [Display definition pop

up.].

CSV file must be formatted as follows:

1st column: ID (character string)

2nd column: Text (character string)

No title row

Precautions

When reading a CSV file for setting, do not specify a CSV file in a folder with the encryption attribute enabled as the file

source or destination.

21 Analyzer21.12 Displaying Data after Running 365

21

21.12 Displaying Data after RunningShows data read by Reader as well as data registered to a dataset as a result of executing an analysis or diagnostic logic in

various forms within a single window.

To use this function, a dataset needs to be configured in advance.

Page 384 Management

Operating procedure

1. Select [Data display after logic execution] [Data display after execution (single)].

2. The dataset set in 'Management' is shown in 'Select dataset'.

To change dataset selection, click 'Select Dataset'.


3. Display the data as follows:

• Waveform display (single section) (Page 368 Waveform display (single section))

• Waveform display (single variable) (Page 369 Waveform display (single variable))

• Waveform display (complex) (Page 370 Waveform display (complex))

• Text display (Page 371 Text display)

• Synchronous display (Page 372 Synchronous display)

• Supplemental data display (Page 373 Displaying supplemental data)

36621 Analyzer21.12 Displaying Data after Running

Screen configurationThe following shows the screen configuration for displaying data after execution.

Displayed items

Item Description

(1) Trend graph (waveform) display screen Shows the waveform.

• The vertical axis shows the value of the selected variable.

• The horizontal axis shows the number of seconds after the start point.

The display status of the waveform currently snapped (grabbed) by the cursor can be switched to

hide.


[Parallel] button Clicking the [Parallel] button arranges multiple variables vertically.

This button turns to yellow in the parallel display.

Clicking it again cancels the parallel display.

[Left] tab/[Right] tab Specify the variable or section to display the waveform.

(2) Select dataset Shows dataset selected in 'Management'.

Page 384 Management

The dataset can be changed by clicking it.


(3) Select variable Clicking it allows for selecting a variable.

[Var.] button Click this button to display the waveform of a specified variable.

Variable name*1 Shows the variable name.

(4) Section Click to switch sections (0 to 2147483647).

[Sctn] button Click this button to display the waveform of a specified section.

Start time*2 Shows the start time of the section.

End time*2 Shows the end time of the section.

[Test display] tab Shows information of the text variable at the cursor position.

Text variable name Shows the text variable name.

Variable Shows the text variable value.

(1)

(2)(3)

(4)

(5)


21

*1 Shown when the [Var.] button is clicked.*2 Shown when the [Sctn] button is clicked.

[Synchronize] tab When the rising or falling time of the waveform is misaligned with that of another waveform,

synchronize them.

In addition, waveforms can be overlapped by resolving misalignment for each step.

Sync state variable name When synchronizing waveforms by specifying a variable name and value, select a step variable.

Sync state variable value When synchronizing waveforms by specifying a variable name and value, specify a step variable

value.

Order for the same variable

value

When the same step variable value occurs repeatedly within a section, specify which one to use (1

to 2147483647).

[Synchronizes with cursor

snap point.] button

Click this button to synchronize waveforms at the point specified by the cursor.

The cursor must be visible in the trend graph (waveform) display screen.

Waveforms are synchronized at the point in the X-axis direction where each waveform first

crosses the Y value specified by the cursor.

[Sync. by val.] button Click this to perform synchronization by using entered value (-2147483648.0000 to

2147483647.0000).

While this button is clicked (shown in yellow), waveforms are synchronized automatically even

when waveforms or sections are switched. However, synchronization at the point specified by the

cursor is disabled.

Clicking this button cancels its selection.

Waveforms are synchronized at the point in the X-axis direction where each waveform first

crosses the entered Y value.

[Supplement] tab Shows information on a single point or between two points on a waveform snapped to the cursor.

V-differ Shows the difference between the first and second cursors (absolute value) in the Y axis.

The value changes only when two cursors are shown.

H-differ Shows the difference between the first and second cursors (absolute value) in the X axis.

The value changes only when two cursors are shown.

Y coordinates -1 Shows the value in the Y axis for the first cursor.

Y coordinates -2 Shows the value in the Y axis for the second cursor.

This item is shown only when two cursors are shown.

Left EQ Shows the fixed string "UNIT".

Right EQ Shows the fixed string "UNIT".

Left alias Unused area.

Right alias Unused area.

Cursor snap time Shows the time of the point snapped by the cursor.

• Upper display area: Time of the point snapped by the first cursor

• Lower display area: Time of the point snapped by the second cursor





Item Description


Waveform display (single section)Shows the waveform for a single section of a selected variable.

Operating procedure

1. Click the [Var.] button and select a variable from the variable name list. (Multiple selections allowed.)

2. Click to change a section to be displayed (0 to 2147483647).

3. By clicking the [Parallel] button while multiple variables are selected, the variables are displayed vertically.*1

*1 Values in Y axis (left axis) and cursor will not be displayed.

By operating the slider of parallel display, the position of displayed waveforms can be shifted.

• Slider is positioned at the center: Waveforms before applying the parallel display are displayed.

• Slider is positioned a location other than the center: Waveforms displayed in parallel are shifted vertically.

When the slider is positioned at the leftmost or rightmost, the up and down of waveforms displayed in parallel are reversed.


21

Waveform display (single variable)The waveform of a single variable in the selected section is shown.

Operating procedure

1. Click the [Sctn] button and select a section. (Multiple selections allowed.)

2. Click to change variable to be displayed.

3. By clicking the [Parallel] button while multiple variables are selected, the variables are displayed vertically.*1

*1 Values in Y axis (left axis) and cursor will not be displayed.

By operating the slider of parallel display, the position of displayed waveforms can be shifted.

• Slider is positioned at the center: Waveforms before applying the parallel display are displayed.

• Slider is positioned a location other than the center: Waveforms displayed in parallel are shifted vertically.

When the slider is positioned at the leftmost or rightmost, the up and down of waveforms displayed in parallel are reversed.


Waveform display (complex)Shows two waveforms by overlapping them.

Operating procedure

1. Select the [Right] tab and configure settings similarly to those on the [Left] tab to display waveforms overlapped.

2. By right-clicking the trend graph (waveform) display screen and selecting [Base] [Clear], the selection of variables in

the [Right] and [Left] tab will be canceled.

3. By selecting [Base] [Clear right axis], the selection of variables in the [Right] tab will be canceled. By selecting [Base]

[Clear left axis], the selection of variables in the [Left] tab will be canceled.

Settings on the [Left Axis] and [Right Axis] tab are reflected on the left and right sides of the Y-axis,

respectively, in the trend graph (waveform) display screen.


21

Text displayShows information of the text variable at the cursor position.

Operating procedure

1. Display the cursor in the trend graph (waveform) display screen.

For showing the cursor, refer to the following:


2. Click in the trend graph (waveform) display screen.

The cursor moves to the point where the clicked point intersects with the waveform. Information of the text variable at that

point is shown in the [Test display] tab.

For the display contents, refer to the following:


3. By right-clicking on the trend graph (waveform) display screen and selecting [Cursor] [Delete], all cursors in a graph

will be deleted.


Synchronous displayA synchronization is performed between different variables selected in the [Left] tab or between the same variables in different

sections of variables selected in the [Left] tab.

When the rising or falling time of the waveform is misaligned with that of another waveform, they can be synchronized.

In addition, waveforms can be overlapped by resolving misalignment for each step.

The following synchronization methods are available:

• Synchronize by a variable name and value

• Synchronizes at cursor snap point

• Synchronize by value

Operating procedure

1. Set items in the [Synchronize] tab.

For details on setting items, refer to the following:


Items to set differs depending on the synchronization method.

• Synchronizing by a variable name and value: Set the settings for "Sync state variable name", "Sync state variable value", or

"Order for the same variable value".

• Synchronizes with cursor snap point.: Click the [Synchronizes with cursor snap point.] button while snapping a waveform

with the cursor.

• Sync. by val.: Enter a value (-2147483648.0000 to 2147483647.0000) and click the [Sync. by val.] button.

2. Waveforms are overlapped according to the specified method.

When waveforms are synchronized by a variable name and value, they are shifted in the X-axis direction to align the step

change times.

When waveforms are synchronized by a point specified by the cursor or by a value, they are shifted in the X-axis direction so

that they pass the same Y value.


21

Displaying supplemental dataShows information at the point or between two points on the waveform snapped to the cursor in the [Supplement] tab.

Operating procedure

1. Display the cursor in the trend graph (waveform) display screen.

For showing the cursor, refer to the following:


2. Information is shown on the [Supplement] tab.

For the display contents, refer to the following:


Hiding a waveformThe display status of the waveform snapped (grabbed) by the cursor can be switched to hide.

Operating procedure

1. Right-click on the display screen and select [Clear snapped waveform].

When redisplaying a waveform, perform the following operations:

• When the [Var.] button has been clicked:

Select the variable corresponding to the hidden waveform. (When hiding a waveform, the corresponding

section will be unselected in the section list.)

• When the [Sctn] button has been clicked:

Select the section corresponding to the hidden waveform. (When hiding a waveform, the corresponding

section will be unselected in the section list.)

37421 Analyzer21.13 Displaying a Result of Simple diagnosis

21.13 Displaying a Result of Simple diagnosisThe diagnosis results of 'SPC', 'multivariate analysis (single central value format)', or 'guard band' in process are displayed.

Page 189 SPC, Page 207 Multivariate Analysis (Single Central Value Format), Page 215 Guard Band Diagnosis

Operating procedure

1. Select [Diagnostics result display] [Simple diagnosis result display].

2. Display the diagnosis result by clicking one of the following buttons:

• [Trend] button (Page 377 Displaying trend chart)

• [Beep] button (Page 378 Beep)

• [Total] button (Page 380 Displaying aggregated data)

• [Log] button (Page 381 Viewing log)

21 Analyzer21.13 Displaying a Result of Simple diagnosis 375

21

Screen configurationThe following shows the screen configuration for displaying the result of simple diagnosis.

For the screen configuration when clicking the [Trend] button, [Beep] button, [Total] button, or [Log] button, refer to:

Displaying trend chart

Page 378 Beep

Page 380 Displaying aggregated data

Page 381 Viewing log

Displayed items

*1 It cannot be selected while the [Beep] button is clicked.

Item Description

Diag-name Shows the name of the simplified diagnosis currently in progress.

For a diagnosis for which a filtering text has been specified, a diagnosis name can be displayed by specifying a filtering

text and its name.

Shows error information related to the selected diagnosis name.

When nothing is selected, error information for all displayed diagnosis name is displayed.

[Trend] button Shows a trend chart.

Page 377 Displaying trend chart

[Beep] button If an error is detected in diagnosis started with simple diagnosis, a log will be displayed and the pop-up window will be

displayed.

Page 378 Beep

[Total] button Selects aggregated data for the selected equipment, unit, or period.

Page 380 Displaying aggregated data

[Log] button Shows the log.


[ALL] button*1 Shows the results of all simple diagnoses.

[GB] button*1 Shows the result of simple diagnosis related to a guard band.

[SPC] button*1 Shows the result of simple diagnosis related to SPC.

[MVA] button*1 Shows the result of simple diagnosis related to multivariate analysis (single central value format).

[Log CSV] button Output a determination result for a section in which the one or more number of diagnoses is displayed in the log screen

and the detailed information to a CSV file.


For the path of a file to be output, up to 255 characters (two characters are used for a double-byte character) including

the extension can be specified.





• [Save] button

• [Read] button

• [Multi] button


• [CV] button

• [FLT] button

• [3D] button


• [Lock] button


• [Max] button

• [Min] button

• [Ave] button

• [Range] button

• [Center] button

• [Sigma] button


• Histogram

• [Var.] tab

• [EQ Def] tab

• Up/Down/Left/Right button


21

Displaying trend chartShows a trend chart.

Operating procedure

1. Click the [Trend] button.

2. A trend chart is shown.

3. Snap the cursor to a section to show the text information related to that section.

While the text information for the section is displayed, click the [Sctn] button to show the section list with the

section selected.


4. Clicking the [Update] button updates the display contents to the data of the latest time.

Item Description

(1) Error count display screen (per

day)

Shows the number of error occurrences per day as a trend chart.

Snap the cursor to a day to show the trend chart of error occurrences per section of that day in the error count display

screen (per section on that day).

(2) Error count display screen (per

section on that day)

Shows the number of error occurrences on the selected day as a trend chart.

(1)

(2)


BeepIf an error is detected in diagnosis started with simple diagnosis (SPC, multivariate analysis (single central value format), or

guard band), a log is displayed and a beep sounds. The pop-up window is also displayed.

Operating procedure

1. Click the [Beep] button.

2. When an error is detected during the simplified diagnosis, a log is shown at the top of the log list. (When "Refreshing ON/

OFF" is set to "ON")

In addition, the beep sound is generated according to the setting of "Beeping".


Item Description

(1) Log list Shows the log related to the error.

Beep STOP Stops the beep sound.

Refreshing ON/OFF Switches ON and OFF of refresh.

When switching to "OFF", updating the list stops and a beep sound stops. In addition, the pop-up window display stops. (Diagnosis

itself continues.)

Beeping Specify the time to sound the beep tone.

The beep tone sounds once when the setting has been changed.

Beep length (msec) Specify the length of the beep sound (1 to 10000 milliseconds).

Tone Specify the tone of the beep.

(1)


21

4. The pop-up window appears at the same time when a log is displayed on the top in the log list.

• Up to 2000 lines are shown for a single log. Starting Analyzer clears the log.

• The pop-up window also appears when a log is selected in the log list.

Precautions

• Error information detected while "Refreshing ON/OFF" is set to "OFF" is not reflected on the log list even when the [Update]

button is clicked.

• If an error is detected during data reading on a first section since an SPC or a multivariate analysis (single central value

format) has been started, a waveform to be diagnosed will not be displayed in the pop-up window (because the waveform

to be displayed is only one).

(1) Shows the variable name.

(2) Shows error occurrence time and error value

(3) Shows the diagnosis name.

(1) (2)

(3)


Displaying aggregated dataDisplays aggregated data of the equipment, unit, and diagnosis name for a period specified for "From" and "To".

Operating procedure

1. Click the [Total] button.

2. Aggregated data of the equipment, unit, and diagnosis name is shown.


When diagnosis is stopped, the diagnosis name disappears. When the [Update] button is clicked in this state,

the count shown for each diagnosis name disappears and the equipment and unit counts are rest to '0'.

Item Description

(1) Equipment list Displays the equipment list.

(2) Error occurrences display

area (per equipment)

Shows error occurrences per equipment.

(3) Unit list When a device is selected in the equipment list, the unit defined for that device is shown.

(4) Error count display area (per

unit)

Shows the number of error occurrences per unit.

(5) Diagnosis name list When a unit is selected from the unit list, the name of diagnosis currently performed on the unit is shown.

(6) Error count display area (per

diagnosis name)

Shows the number of error occurrences per diagnosis name.

(1)

(3)

(5)(2)

(4)

(6)


21

Viewing logShows the log.

Operating procedure

1. Click the [Log] button.

2. The log is displayed.


When diagnosis is stopped, "Error Count" and "Diag-num." on the error information row in the log list changes

to '0'. When diagnosis is resumed, data before it was stopped is shown again.

Precautions

If an error is detected during data reading on a first section since a multivariate analysis (single central value format) and SPC

have been started, a waveform to be diagnosed will not be displayed in 'Waveform when an error occurs' (because the

waveform to be displayed is only one).

Item Description

(1) Log list Logs are displayed.

(2) Diagnosis contents The list of diagnoses performed in a section selected in the log list is displayed.

(3) Detailed information Detailed information is displayed when selecting a row that "NG" is selected for "Judgt." in the diagnosis contents.

(4) Waveform when an error

occurs

A waveform when an error occurs is displayed when selecting a row that "NG" is selected for "Judgt." in the diagnosis

contents.

For a guard band, the horizontal axis of a waveform represents the section length. For multivariate analysis (single central

value format) and SPC, it represents past sections (up to 30 sections).

(1)

(3) (4)

(2)

38221 Analyzer21.14 Displaying a Diagnostic Logic Result

21.14 Displaying a Diagnostic Logic ResultShows the log written by the logics listed below.

Operating procedure

1. Select [Diagnostics result display] [Diagnostic logic result display].

Logs are shown in the log list.

2. The log is updated at an interval set for "Refreshing cycle (sec)", and the latest pop-up window appears at the same time.

• Up to 1000 logs are displayed.

• The pop-up window is also displayed when selecting a log in the log list.

• The contents displayed in "Item" excluding a text variable and a diagnosis name and "Value" corresponding

to "Item" in the pop-up window are displayed in "Item" and "Value" in the screen of 'Diagnostic logic result

display'.

Logic Description Reference

Enhanced trace GB Log when a guard band error was detected. Page 320 Enhanced trace GB

SPC diagnostics Log when an error was detected with the SPC rule. Page 330 SPC diagnostics

Write log and error notification Output contents configured by the 'Write log and error notification' logic. Page 344 Write log and error notification

(1) Shows a comment entered in the configuration item "Comment" of the primitive 'Enhanced trace GB' or 'Write log and error notification'. For the primitive

'SPC diagnostics', the SPC rule that detected an error first among the SPC rules on the list is shown.

(2) Shows the logic comment entered in "Logic comment" of 'Logic editing'.

(1)(2)

21 Analyzer21.14 Displaying a Diagnostic Logic Result 383

21

Screen configurationThe following shows the screen configuration for displaying the diagnostic logic result.

Displayed items

Item Description

(1) Log list An output log is displayed.

Time of occurrence Shows the time when the log occurred.

Equipment Shows the "UNIT".

Diagnostics variable Shows the diagnosis target variable name.

Comment For the primitive 'Enhanced trace GB' or 'Write Log and Notify Error', text entered in the

setting item "Comment" is shown.

For the primitive 'SPC diagnostics', the SPC rule among the SPC rules on the list that

detected an error first is shown.

Item The contents displayed in "Item" in the pop-up window excluding a text variable and a

diagnosis name are displayed.

Value The contents displayed in "Value" corresponding to "Item" in the pop-up window

excluding a text variable and a diagnosis name are displayed.

Refreshing cycle (sec) Set the interval to update a log to the latest one (10 to 3600 seconds).

It is set to 60 seconds by default.

Refreshing ON/OFF Switches ON and OFF of refresh.

The default is "ON".

When set to "OFF", the log list is not automatically updated at the interval specified in

"Refreshing ON/OFF".

(1)

38421 Analyzer21.15 Management

21.15 ManagementConfigures dataset settings.

When analyzing data read by Reader or data waveforms after logic execution in detail in 'Data display after execution' or

cooperating with Monitor, dataset settings must be configured in advance.


Operating procedure

1. Select [Management] [Dataset].

2. Register the dataset.

Page 387 Registering dataset

3. Set a period and section according to the screen configuration.


4. Data which is no longer necessary can be deleted from the dataset setting.

Page 388 Deleting data from dataset

5. A dataset which was registered by mistake or is no longer necessary can be canceled.

Page 389 Canceling dataset

21 Analyzer21.15 Management 385

21

Screen configurationThe following shows the screen configuration for setting a dataset.

Displayed items

Item Description

From/To Specify the period for data to display.

To display data for the period set by this item, click the [Select period] button and select "Specify

date".

[Select period] button Select the period for data to display from the following:

• 10 days recently



• Specify date

• Release specification

When the period is specified, the background of "From" and "To" becomes blue.

When the period is not specified or "Release specification" is selected, the background of "From"

and "To" becomes white and the entire period is shown.

[Update dataset] button If the up-to-date data read by Reader is not shown, click to update display.

Unit Shows all units registered in the specified folder.

[Period] tab Period When a unit is selected in "Unit", a data period registered in the unit is shown, one day per one line.

Select the period shown on related screens.

[FLT] tab Filters the section shown using a text variable value.

Sections filtered by the text variable values selected in "Text" on the [1] to [5] tabs are shown in

"Sctn".

Filtering type Select "AND" or "OR".

• AND: Filters sections into those which satisfy all of the text variable values selected in "Text" on

the [1] to [5] tabs.

• OR: Filters sections into those which satisfy any one of the text variable values selected in "Text"

on the [1] to [5] tabs.

Text name Select (up to five) text variable names used to filter sections.

[1] to [5] tabs Text variable values selected in "Text name" on the [FLT] tab are shown in order from the [1] to [5]

tabs.

Text Shows the text variable value.

# Shows the number of items for each text variable value.

Sctn Shows sections in the period selected in "Period".

The section name indicates the start time extracted by Reader, or the start time of the text variable.

When text variables are selected in "Text name" on the [FLT] tab, the text variable values are shown

in order from the "1" column.

[logic] tab logic Shows the logic run during the period selected in "Period" and on the section selected in "Sctn".

Selecting a logic shows the execution result of that logic.

A logic can be executed for a selected period or section.


[Var.] tab Variable name Shows variable information (variable name, data type, description, and ID) included in data

registered to the unit.

When a logic is selected in "logic", names of variables output by the logic are also shown.

[Length] tab Start time Shows the start time of each section registered to the unit.

Length Shows the section length (in seconds) in the section data registered to the unit.

Interval An interval (second) from the end time of the previous section to that of this section is displayed.

Switch Pattern Switches to a dataset pattern registered in advance.

The upper-level folder name of a folder set for 'RSD dataset folder' in Reader is automatically

registered as a pattern name when restarting Analyzer. By selecting this pattern name in this item, a

dataset with all units included in 'RSD dataset folder' are expanded is displayed

A pattern can be registered by selecting items for "Unit", "Period", and "Sctn" and right-clicking on

the display area of "Dataset (DS)" and selecting [Register pattern].

Dataset (DS) Shows the list of dataset names.

The dataset name is shown in the format 'unit name_UNIT'.

The dataset selected in this item is shown on related screens.

Item Description


21

Registering datasetRegister a dataset.

Operating procedure

1. Select a blank line in "Dataset (DS)".

2. Right-click in "Unit" and select [Specify new folder].

3. Specify the 'UNIT' folder specified in 'RSD dataset folder' in Reader.


4. The dataset of the unit read by Reader is registered.


Deleting data from datasetDelete data from the dataset settings.

The following data can be deleted.

• Unit data (Page 388 Deleting unit data)

• Unit data within a section (Page 388 Deleting unit data within period)

• Data generated by a logic (Page 388 Deleting data generated by logic)

To delete data, selecting the checkbox of "Enable data deletion from the data set screen." in 'Data management setting' is

required in advance.

Data Management Setting

• A unit and a logic itself are not deleted.

• Data is deleted per period. Deletion per section is not available.

■ Deleting unit data

Operating procedure

1. Right-click a unit in "Unit", then select [Delete unit data].

Data of the selected unit is deleted.

■ Deleting unit data within period

Operating procedure

1. Right-click a data period in "Period" and select [Delete period].

Data of the unit within the selected period is deleted.

■ Deleting data generated by logic

Operating procedure

1. Select a logic in "Logic", then right-click in "Period" and select [Delete period].

Output data of the selected logic is deleted. Collected data is not deleted.


21

Canceling datasetCancel a dataset which was registered by mistake or is no longer necessary.

Operating procedure

1. Right-click a dataset in "Dataset (DS)" and select [Cancel Dataset].

The selected dataset is canceled.

2. To cancel all datasets in batch, right-click a dataset in "Dataset (DS)" and select [Cancel All datasets].

[Cancel Dataset] or [Cancel All datasets] does not delete the data itself but just deletes dataset definition(s).

39021 Analyzer21.16 Data Management Setting

21.16 Data Management SettingConfigures settings related to data management.

Operating procedure

1. Select [Option] [Setting].

2. Select "Data management setting".


Precautions

Analyzer must be restarted to reflect the changes in 'Data management setting'.

Item Description

Enable data deletion from the data set screen. A unit and its data selected for "Unit" in the operation screen of 'Management', or data of a

period selected for "Period" can be deleted.

Page 384 Management

Update period list automatically. It must be selected.

Display the instrument difference on the display menu screen. By selecting this checkbox, shows the [EQ Def] tab is displayed in the operating screens of

'Read data display' or 'Read data statistics display' and the data of different units can be

displayed simultaneously.



21 Analyzer21.17 Waveform Display Setting 391

21

21.17 Waveform Display SettingConfigures settings related to waveform display.

Operating procedure


2. Select "Waveform display setting".


Item Description

Graph background color Sets the background color of a graph displayed in the screen.

Trend graph display maximum point number of display menu,

simplified analysis / simplified diagnosis menu screen

Sets the upper limit of the number of points in a trend graph displayed in the operation

screen for the [Disp] menu ('Read data', 'Read data statistics', and 'Read data frequency')

and the [Simple analysis - Simple diagnosis] menu ('SPC', 'MVA (Single central value

format)', and 'Guard band').




Page 189 SPC



Displayed data display and read data screen Maximum number

of points displayed

Sets the upper limit of the number of points displayed in the operating screen of 'Data

display after logic execution' and 'Read data display'.



Coefficient multiplied by X axis in line chart. Specifies how much the time axis of the original data should be multiplied in the operation

screen of 'Data display after logic execution'.


Number of bins in the histogram of the display menu, simplified

analysis / simplified diagnosis menu screen

Sets the number of histogram bins (10 to 100) in the operation screen for the [Disp] menu

('Read data', 'Read data statistics', and 'Read data frequency') and the [Simple analysis -

Simple diagnosis] menu ('SPC', 'MVA (Single central value format)', and 'Guard band').




Page 189 SPC



39221 Analyzer21.17 Waveform Display Setting

Precautions

• Analyzer must be restarted to reflect the changes to the waveform display settings.

• When the upper limit of the number of display points is exceeded, an error is output and chart display is interrupted.

21 Analyzer21.18 Logic Setting 393

21

21.18 Logic SettingConfigures settings related to logic.

Operating procedure


2. Select "Logic setting".


*1 Do not specify a folder with the encryption attribute enabled.*2 Set it so that the path length of a file to be saved does not exceed 259 characters. The maximum path length of a pop up error log is as

follows:(Set folder)\ + 49 characters (if the logic name is 32 characters long)

Precautions

Analyzer must be restarted to reflect the changes to the logic settings.

Item Description

Number of batch sections Do not use this.

Specify the folder that saves a pop up error log.*1*2 Sets save destination (up to 256 characters) for a log when the pop-up window appears by

performing a logic manually in 'Logic editing'.


39421 Analyzer21.19 Programmable Controller CPU Cooperation Function

21.19 Programmable Controller CPU Cooperation Function

This function is for diagnosing data on a programmable controller CPU in real time using the result of offline analysis created

in Real-time Statistic Diagnosis Tool.

Page 215 Guard Band Diagnosis, Page 243 Multiple Regression Analysis (LMR), Page 250 Maharanobis - Taguchi

method (MT)

Guard band diagnosisTo use the result of offline analysis created in a guard band diagnosis in a programmable controller CPU, follow the procedure

below.

Create a guard band. (Page 220 Creating guard band)

Save the guard band in the programmable controller format (.PGB). (Page 221 Saving the guard band)

Write the guard band file (.PGB) to an SD memory card. (Page 397 Writing to an SD memory card)

Execute the Real-time Data Analyzer data linkage FB. (Page 398 Execution of Real-time Data Analyzer data linkage

FB)

Execute the data analysis FB (guard band diagnosis FB). (Page 400 Executing the data analysis FB (guard band

diagnosis FB))

Precautions

• For the temporary license, a guard band cannot be saved in the programmable controller format.

• To save a guard band in the programmable controller format, the data interval and length (number of units of data) of the

guard band need to be the same as that of the waveform to be diagnosed by an FB. In that case, perform the following

operations.

Set the sampling interval of a CSV file to be read in Reader to the same as the collection interval of the waveform data to be

diagnosed by an FB, and write it to a file register (ZR).

(1) Industrial PC


(3) SD memory card

A: Guard band

B: Guard band file (.PGB)

(3)

(1)

(2)

AA BBB� ��

BB

FB

��

��

21 Analyzer21.19 Programmable Controller CPU Cooperation Function 395

21

Configure the settings for sections or extended sections in Reader so that the length of each section becomes constant,

and then set the length (number of units of data) of a guard band to be created to the same as the data length (number of

units of data) of the waveform to be diagnosed by the FB.

(Example) Prepare variables to detect the start and end of a section as the collection targets, and extract sections so that the

length of each section between its start and end becomes constant.


• To prepare the following variables A and B for detecting the start and end of a section, configure the section start and end

conditions by setting "Numeric" in "Starting condition type"/"End condition type", "UP(Less than TO greater than or equal)"

in "Operator", and "0.5" in "Comparison value". (Page 137 Section start conditions, Page 140 Section end conditions)

Use data whose sampling intervals are constant and which have no data missing (data with no need for interpolation) for

the data of the CSV file to be read in Reader.

• A guard band can be saved in the programmable controller format when the following conditions are satisfied. When is

not satisfied, the value of 25th digit will be rounded off. When neither nor is satisfied, an error will occur.

The number of digits (including signs ('+' or '-') and a decimal separator) is less than 24.

The number of digits for the integer part of a single-precision real number data is less than 16.

(1) Variable A (for section start detection)

(2) Variable B (for section end detection)

(3) Extraction target variable

(4) Extraction section

(a) Set conditions so that the time between the rises of variables A and B becomes constant.

(3)

(1)

(2)

(a)0

0.5

0

0.5

(4)


21

Writing to an SD memory cardWrite a guard band file (.PGB) saved in the programmable controller format to an SD memory card.

The following shows the writing method using GX Works3.

Operating procedure

1. Insert an SD memory card into a programmable controller CPU.

2. Start GX Works3, and create a project.

3. Select [Online] [User Data] [Write].

4. Specify "Write TargetFolder" and "SD Write TargetFolder."

• Write TargetFolder: Specify the folder where a guard band file (.PGB) is stored.

• SD Write TargetFolder: Specify the folder to write the guard band file (.PGB).

5. Select the checkbox of a guard band file (.PGB) to be written, and click the [Execute] button.


Execution of Real-time Data Analyzer data linkage FBStore data of the guard band file (.PGB), which is in an SD memory card, in a file register (ZR) of a programmable controller

CPU by the Real-time Data Analyzer data linkage FB (M+RDAData_ImportGB_R).

For details on the Real-time Data Analyzer data linkage FB, refer to the following:

MELSEC iQ-R Real-time Data Analyzer Data Linkage Function Block Library Reference

To acquire the Real-time Data Analyzer data linkage FB (M+RDAData_ImportGB_R), please contact your

local Mitsubishi Electric sales office or representative.

■ Program exampleThe following shows a program example to execute the Real-time Data Analyzer data linkage FB

(M+RDAData_ImportGB_R).

■ Settings of the Real-time Data Analyzer data linkage FBThe following shows the settings of Real-time Data Analyzer data linkage FB (M+RDAData_ImportGB_R) used in the

program example.

Variable name Name Setting value Description

i_udReadNum Number of reading

points

K1000 Specify the number of points to be used from the waveforms

output to the guard band file.

To check the number of guard band points, refer to the following:

Page 220 Creating guard band

i_udStartPosition Start position of reading K1 Specify the start point of data to be used from data of the

waveforms output in a guard band file. (When using data of the

start and succeeding points, specify '1'.)

i_udLowerLimitAddr Check value (lower limit

value) address

K0 Specify the start address of a file register (ZR) to store the check

value (lower limit value).

i_udUpperLimitAddr Check value (upper limit

value) address

K2000002 Specify the start address of a file register (ZR) to store the check

value (upper limit value).

i_s59FileName File name 'GB_test.PGB' Specify the file name of a guard band file.

i_s59FolderPath Folder Path '/test/' Specify the folder path on an SD memory card where a guard band

file is stored. (Use '\' or '/' for a delimiter of the path.)


21

■ LabelsThe following shows the labels used in the program example.

• External input

• External output

Precautions

Insert the SD memory card where the guard band file (.PGB) output in Real-time Data Analyzer is stored into a programmable

controller CPU.

Label name Description

bImpGB_EN Execution command for the Real-time Data Analyzer data linkage FB


bImpGB_ENO Execution status of the Real-time Data Analyzer data linkage FB

bImpGB_OK Normal completion of the Real-time Data Analyzer data linkage FB

bImpGB_Err Error completion of the Real-time Data Analyzer data linkage FB

uImpGB_ErrId Error code when the Real-time Data Analyzer data linkage FB is completed with an error

udImpGB_ResRdNum Number of data points read when the Real-time Data Analyzer data linkage FB is normally completed

udImpGB_AllDNum Number of all data points for a guard band file


Executing the data analysis FB (guard band diagnosis FB)Execute the data analysis FB (M+DataAnalysisPro_AryBoundCompareTest_R).

For details on the data analysis FB, refer to the following:

MELSEC iQ-R Data Analysis Function Block Library Reference (Application)

To acquire the data analysis FB (M+DataAnalysisPro_AryBoundCompareTest_R), please contact your local

Mitsubishi Electric sales office or representative.

■ Program exampleThe following shows a program example to execute the data analysis FB (M+DataAnalysisPro_AryBoundCompareTest_R).

■ Settings of the data analysis FBThe following shows the settings of data analysis FB (M+DataAnalysisPro_AryBoundCompareTest_R) used in the program

example.

*1 Specify the same value as 'o_udResultReadNum' of M+RDAData_ImportGB_R.*2 Specify '2' (single-precision real number) for the data type.*3 Specify the same value as 'i_udLowerLimitAddr' of M+RDAData_ImportGB_R.*4 Specify the same value as 'i_udUpperLimitAddr' of M+RDAData_ImportGB_R.


i_udSize Number of data points udImpGB_RdNum*1 Specify the number of data points for waveforms to be diagnosed.

i_udWaveDataAddr Waveform data start

address

K4000004 Specify the start address of the file register (ZR) where the

waveform data to be diagnosed is stored.

i_uWaveDataType Waveform data type

selection

K2*2 Specify the data type of waveform data.

i_udLowerLimitAddr Check value (lower limit

value) address

K0*3 Specify the address of the file register (ZR) where the check value

(lower limit value) is stored.

i_udUpperLimitAddr Check value (upper limit

value) address

K2000002*4 Specify the address of the file register (ZR) where the check value

(upper limit value) is stored.

i_udConsecutivePoints Number of consecutive

excess points

K5 Specify the number of points to define the timing of an NG

determination. It is determined as NG when the range between the

lower and upper limit values is exceeded continuously by the

specified number of points.

i_uDecimalPlaces Number of significant

decimal places

K6 Specify the number of significant digits after the decimal point.


21


• External input

• External output


bAryB_EN Execution command for the data analysis FB


bAryB_ENO Execution status of the data analysis FB

bAryB_OK Normal completion of the data analysis FB

bAryB_Res Check result of the data analysis FB

bAryB_Err Error completion of the data analysis FB

uAryB_ErrId Error code when the data analysis FB is completed with an error


Multiple regression analysisTo use the result of offline analysis created in a multiple regression analysis to a programmable controller CPU, follow the

procedure below.

Perform correlation analysis. (Page 235 Correlation analysis by using a correlation graph)

Perform recurrence calculation. (Page 247 Recurrence calculation)

Output the unit space data of the criterion matrix in the programmable controller format (.PMR). (Page 403 External

output)

Write a multiple regression analysis file (.PMR) to an SD memory card.( Writing to an SD memory card)


FB)

Execute the multiple regression analysis FB. (Page 407 Execution of multiple regression analysis FB)

(1) Industrial PC


(3) SD memory card

A: Correlation analysis result

B: Unit space data

C: Multiple regression analysis file (.PMR)

(3)

(1)

(2)

AA CCC� ��

CC

FB

��

��BB


21

External outputOutput the unit space data of the criterion matrix in the format available in a programmable controller.

The unit space data includes the following:

Operating procedure

Click the [External output] button in the screen shown in the following section.


The unit space data is output.

■ Output formatThe following table shows the format for outputting the unit space data.


automatically added and the data is saved.

Precautions

• For the temporary license, the [External output] button cannot be clicked.

• Unit space data of the criterion matrix that is output when clicking the [External output] button may not be the same value as

the data output by the FB (M+DataStatisticsPro_LinMultRg_ED_R) which generates unit space data.

• The available data type in a programmable controller is the single-precision real number type. When saving in the

programmable controller format and if the value of unit space data of criterion matrix falls below the range that can be

represented in single-precision real numbers, the value will be rounded so as to fall within the range*1 and saved. In

addition, when the range that can be represented in single precision real numbers is exceeded, an error will occur.

*1 It will be rounded off.

• The maximum number of units of unit space data for the criterion matrix is 64 excluding objective variables. An error occurs

if there are more than 64 units.

• The value of a target variable for recurrence calculation in a correlation matrix (partial matrix) needs to be set within the

range that can be represented in single-precision real numbers.

• The maximum number of target variables for recurrence calculation in a correlation matrix (partial matrix) is 64 excluding

objective variables.

Item name Data type Number of units of data Corresponding item of the FB

Number of target variables for recurrence

calculation

Word [unsigned] 1 Number of signal data items

Coefficient Single-precision

real number

Value of 'Number of target variables for recurrence

calculation'

Coefficient

Constant Single-precision

real number

1 Constant


Programmable controller format (.PMR) This format is available in a programmable controller.


Writing to an SD memory cardWrite a multiple regression analysis file (.PMR) saved in the programmable controller format to an SD memory card.


Operating procedure





• Write TargetFolder: Specify the folder where a multiple regression analysis file (.PMR) is stored.

• SD Write TargetFolder: Specify the folder to write the multiple regression analysis file (.PMR).

5. Select the checkbox of a multiple regression analysis file (.PMR) to be written, and click the [Execute] button.


21

Execution of Real-time Data Analyzer data linkage FBStore data of a multiple regression analysis file (.PMR), which is in an SD memory card, in a file register (ZR) of a

programmable controller CPU by the Real-time Data Analyzer data linkage FB (M+RDAData_ImportMR_R).



To acquire the Real-time Data Analyzer data linkage FB (M+RDAData_ImportMR_R), please contact your



(M+RDAData_ImportMR_R).

■ Settings of the Real-time Data Analyzer data linkage FBThe following shows the settings of Real-time Data Analyzer data linkage FB (M+RDAData_ImportMR_R) used in the

program example.


• External input

• External output

Precautions


i_uReadItemLimit Maximum number of

items to read

K64 Specify a value equal to or larger than the number of target

variables for 'recurrence calculation' when creating a multiple

regression analysis file.

i_udInMCoefAddr Coefficient m area start

address

K100 Specify the start address of the file register (ZR) to store the

coefficient m. (The data is output in the single-precision real

number type.)

i_s59FileName File name 'PmrFi.PMR' Specify the file name of the multiple regression analysis file.

i_s59FolderPath Folder Path 'PMRs' Specify the folder path on an SD memory card where a multiple

regression analysis file is stored. (Use '\' or '/' for a delimiter of the

path.)


bImpPMR_EN Execution command for the Real-time Data Analyzer data linkage FB


bImpPMR_ENO Execution status of the Real-time Data Analyzer data linkage FB

bImpPMR_OK Normal completion of the Real-time Data Analyzer data linkage FB

bImpPMR_Err Error completion of the Real-time Data Analyzer data linkage FB

uImpPMR_ErrId Error code when the Real-time Data Analyzer data linkage FB is completed with an error

eBConst_val Constant stored in the multiple regression analysis file

uDataItem Number of target variables for 'recurrence calculation' when creating the multiple regression analysis file


Insert the SD memory card where the multiple regression analysis file (.PMR) output in Real-time Data Analyzer is stored into

a programmable controller CPU.


21

Execution of multiple regression analysis FBExecute the multiple regression analysis FB (M+DataStatisticsPro_LinMultRgCalc_E_R).

For details on the multiple regression analysis FB, refer to the following:

MELSEC iQ-R Statistical Analysis Function Block Library Reference (Application)

To acquire the multiple regression analysis FB (M+DataStatisticsPro_LinMultRgCalc_E_R), please contact

your local Mitsubishi Electric sales office or representative.

■ Program exampleThe following shows a program example to execute the multiple regression analysis FB

(M+DataStatisticsPro_LinMultRgCalc_E_R).

■ Settings of the multiple regression analysis FBThe following shows the settings of multiple regression analysis FB (M+DataStatisticsPro_LinMultRgCalc_E_R) used in the

program example.

*1 Specify the same value as 'o_uDataItem' of M+RDAData_ImportMR_R.*2 Specify '2' (single-precision real number) for the data type.*3 Specify the same value as 'i_udInMCoefAddr' of M+RDAData_ImportMR_R.*4 Specify the same value as 'o_eBConst' of M+RDAData_ImportMR_R.


i_uDataItem Number of signal data

items

uDataItem*1 Specify the number of explanatory variable x.

(Example) If the explanatory variable x to be calculated (X) equals

X0, X1, ..., and Xn-1, specify 'n.'

i_uDataType Data type selection K2*2 Specify the data type of waveform data.

i_udInXExpVarAddr Start address of

predictor variable x

storage destination

K2000 Specify the start address of the file register (ZR) where the

explanatory variable x is stored.

Data for the 'number of signal data items' is stored in the file

register (ZR).

The data is stored in the data type same as the one selected for

the 'number of signal data items.'

For the explanatory variable, store the statistics calculated with the

same method as the one used for generating unit space data.

i_udInMCoefAddr Start address of

coefficient m storage

location

bImpPMR_Adr*3 Specify the start address of the file register (ZR) where the

coefficient m is stored.

Data for the 'number of signal data items' is stored in the file

register (ZR).

The data is stored in the single-precision real number type.

i_eBConst Constant b eBConst_val*4 Specify the constant b.



• External input

• External output


bAryB_EN Execution command for the multiple regression analysis FB


bAryB_ENO Execution status of the multiple regression analysis FB

bAryB_OK Normal completion of the multiple regression analysis FB

bAryB_Err Error completion of the multiple regression analysis FB

uAryB_ErrId Error code when the multiple regression analysis FB is completed with an error

eAryYObj Response variable y


21

Maharanobis-Taguchi methodTo use the result of offline analysis created by Mahalanobis-Taguchi method in a programmable controller CPU, following the

procedure below.

Perform correlation analysis. (Page 235 Correlation analysis by using a correlation graph)

Calculate MT distance. (Page 252 MT distance calculation)

Output the unit space data of the criterion matrix in the programmable controller format (.PMT). (Page 410 External

output)

Write an MT method file (.PMT) to an SD memory card.( Writing to an SD memory card)


FB)

Execute the statistical analysis FB (MT method FB). (Page 414 Execution of statistical analysis FB (MT method FB))

(1) Industrial PC


(3) SD memory card

A: Correlation analysis result

B: Unit space data

C: MT method file (.PMT)

(3)

(1)

(2)

AA CCC� ��

CC

FB

��

��BB


External outputOutput the unit space data of the criterion matrix in the format available in a programmable controller.

The unit space data includes the following:

Operating procedure

Click the [External output] button in the screen shown in the following section.


The unit space data is output.

■ Output formatThe following table shows the format for outputting the unit space data.


automatically added and the data is saved.

Precautions

• For the temporary license, the [External output] button cannot be clicked.

• Unit space data of the criterion matrix that is output when clicking the [External output] button may not be the same value as

the data output by the FB (M+DataStatisticsPro_MTUnit_R) which generates unit space data.

• The available data type in a programmable controller is the single-precision real number type. When saving in the

programmable controller format and if the value of unit space data of criterion matrix falls below the range that can be

represented in single-precision real numbers, the value will be rounded so as to fall within the range*1 and saved. In

addition, when the range that can be represented in single precision real numbers is exceeded, an error will occur.

*1 It will be rounded off.

• When outputting data in the programmable controller format, set the value of a target variable for MT distance calculation in

a correlation matrix (partial matrix) within the range that can be represented in single-precision real numbers.

Item name Data type Number of units of data Corresponding item of the FB

Number of target variables for MT distance

calculation

Word [unsigned] 1 Number of signal data items

Inverse matrix of correlation matrix Single-precision

real number

Value of 'Number of target variables for MT distance

calculation' 2

Inverse matrix

Average value per item Single-precision

real number


calculation'

Average value

Standard deviation per item Single-precision

real number


calculation'

Standard deviation


Programmable controller format (.PMT) This format is available in a programmable controller.


21

Writing to an SD memory cardWrite an MT method file (.PMT) saved in the programmable controller format to an SD memory card.


Operating procedure





• Write TargetFolder: Specify the folder where an MT method file (.PMT) is stored.

• SD Write TargetFolder: Specify the folder to write the MT method file (.PMT).

5. Select the checkbox of an MT method file (.PMT) to be written, and click the [Execute] button.


Execution of Real-time Data Analyzer data linkage FBStore data of an MT method file (.PMT), which in an SD memory card, in a file register (ZR) of a programmable controller CPU

by the Real-time Data Analyzer data linkage FB (M+RDAData_ImportMT_R).



To acquire the Real-time Data Analyzer data linkage FB (M+RDAData_ImportMT_R), please contact your



(M+RDAData_ImportMT_R).

■ Settings of the Real-time Data Analyzer data linkage FBThe following shows the settings of Real-time Data Analyzer data linkage FB (M+RDAData_ImportMT_R) used in the

program example.


• External input

Variable Name Name Setting value Description

i_uReadItemLimit Maximum number of

items to read

K300 Specify the value equal to or larger than the number of target

variables for 'MT distance calculation' when creating the MT

method file.

i_udOutAveAddr Output average start

address

K0 Specify the start address of a file register (ZR) to store the average

value. (The data is output in the single-precision real number

type.)

i_udOutSDAddr Output SD start address K700 Specify the start address of a file register (ZR) to store the

standard deviation. (The data is output in the single-precision real

number type.)

i_udOutInvAddr Output invmatrix start

address

K1400 Specify the start address of a file register (ZR) to store the inverse

matrix. (The data is output in the single-precision real number

type.)

i_s59FileName File name 'MT_method_param.PMT' Specify the file name of an MT method file.

i_s59FolderPath Folder path '/test/' Specify the folder path on a SD memory card where an MT method

file is stored. (Use '\' or '/' for a delimiter of the path.)


bImpMT_EN Execution command for the Real-time Data Analyzer data linkage FB


21

• External output

Precautions

Insert the SD memory card where the MT method file (.PMT) output in Real-time Data Analyzer is stored into a programmable

controller CPU.


bImpMT_ENO Execution status of the Real-time Data Analyzer data linkage FB

bImpMT_OK Normal completion of the Real-time Data Analyzer data linkage FB

bImpMT_Err Error completion of the Real-time Data Analyzer data linkage FB

uImpMT_ErrId Error code when the Real-time Data Analyzer data linkage FB is completed with an error

uImpMT_NormItemNum Number of target variables for 'MT distance calculation' when creating the MT method file


Execution of statistical analysis FB (MT method FB)Execute the statistical analysis FB (M+DataStatisticsPro_MTMD_R).

For details on the statistical analysis FB, refer to the following:

MELSEC iQ-R Statistical Analysis Function Block Library Reference (Application)

To acquire the statistical analysis FB (M+DataStatisticsPro_MTMD_R), please contact your local Mitsubishi

Electric sales office or representative.

■ Program exampleThe following shows a program example to execute the statistical analysis FB (M+DataStatisticsPro_MTMD_R).

■ Settings of the statistical analysis FBThe following shows the settings of statistical analysis FB (M+DataStatisticsPro_MTMD_R) used in the program example.


i_uSigItemNum Number of signal data

items

uImpMT_NormItemNum*1 Specify the number of target variables for 'MT distance calculation'

when creating the MT method file.

i_uSigSampleNum Number of signal data

samples

K100 Specify the number of signal data samples.

i_uDataType Data type selection K2*2 Specify the data type of signal data.

i_udInSigAddr Start address of signal

data storage destination

K300000 Specify the start address of the file register (ZR) where the signal

data for calculating the Mahalanobis distance is stored.

To the file register (ZR), store the data for 'Number of signal data

items' 'Number of signal data samples' in the single-precision

real number type.

For the signal data, store the statistics calculated with the same

method as the one used for generating unit space data.

i_udTmpMatrix Start address of an area

for matrix operation

K500000 Specify the start address of a file register (ZR) in the area for

matrix operation.

i_udInAveAddr Start address of average

value storage

destination

K0*3 Specify the start address of the file register (ZR) where the

average value of normal data is stored.

i_udInSDAddr Start address of

standard deviation

storage destination

K700*4 Specify the start address of the file register (ZR) where the

standard deviation of normal data is stored.


21

*1 Specify the same value as 'o_uNormItemNum' of M+RDAData_ImportMT_R.*2 Specify '2' (single-precision real number) for the data type.*3 Specify the same value as 'i_udOutAveAddr' of M+RDAData_ImportMT_R.*4 Specify the same value as 'i_udOutSDAddr' of M+RDAData_ImportMT_R.*5 Specify the same value as 'i_udOutInvAddr' of M+RDAData_ImportMT_R.


• External input

• External output

*1 It is stored when 'SN ratio (larger-is-better characteristic) output enable/disable of signal data MD' is ON.

i_udInInvAddr Start address of inverse

matrix storage

destination

K1400*5 Specify the start address of the file register (ZR) where the inverse

matrix of the correlation matrix of a unit space is stored.

i_udOutMDAddr Start address of signal

data MD output

destination

K200000 Specify the start address of a file register (ZR) to store an output

result (Mahalanobis distance of signal data).

To the file register (ZR), the data for 'Number of signal data

samples' is stored in the single-precision real number type.

i_udOutContAddr Start address of

contribution level output

destination

K210000 Specify the start address of a file register (ZR) to store an output

result (contribution level).

The contribution level is stored to the file register when

'Contribution level output enable/disable of the statistical analysis

FB' is ON. To the file register (ZR), the data for 'Number of signal

data items' 'Number of signal data samples' is stored in the

single-precision real number type.


bMTMD_EN Execution command for the statistical analysis FB

bMTMD_ConLev Contribution level output enable/disable of the statistical analysis FB

bMTMD_SN SN ratio (larger-is-better characteristic) output enable/disable of signal data MD of the statistical analysis FB


bMTMD_ENO Execution status of the statistical analysis FB

bMTMD_OK Normal completion of the statistical analysis FB

bMTMD_Err Error completion of the statistical analysis FB

uMTMD_ErrId Error code when the statistical analysis FB is completed with an error

eMTMD_lgeRespSN SN ratio (larger-is-better characteristic) of signal data MD*1


41622 Monitor

22 Monitor

This software displays the diagnosis status and a diagnosis result, and outputs a response data file when an error occurs in

diagnosis.

For the content output in a response data file, refer to the following:


In addition, the following operations can be performed in this software:

• Outputting a log when a diagnosis is performed

• Outputting a diagnosis result detail information file when an error occurs in a simple diagnosis

For the content output in a diagnosis result detail information file, refer to the following:


The following shows the procedure to operate Monitor.

Operating procedure

1. Set the settings for cooperating with an external tool such as Edgecross Basic Software.

Page 419 Monitor Setting

2. Perform operations for cooperating with Edgecross Basic Software.

Page 421 Starting Monitor

22 Monitor22.1 Screen Configuration 417

22

22.1 Screen ConfigurationThe following shows the screen configuration of Monitor.

Displayed items

Item Description

[DIAG.] button Click this to perform a diagnosis in Monitor.

[Setting] button Click this to set the settings for Monitor.

Unit comment Unused area.

Logic comment A logic comment set for 'logic editing' in Analyzer is displayed.


(1) Unit information list Information on a unit defined in Reader is displayed.

Unit A unit name defined in Reader is displayed.

State "Diagnosing" is displayed.

Last collection time The latest time when a CSV file of a selected unit is read in Reader is displayed.

(2) Logic information list Information on a logic of a unit selected in the unit information list is displayed.

logic A logic name defined in a unit selected in the unit information list is displayed.

A logic indicated as "Monitor" is regarded as simple diagnosis (SPC, multivariate analysis

(single central value format) and guard band) in Analyzer.

Page 189 SPC, Page 207 Multivariate Analysis (Single Central Value Format), Page 215

Guard Band Diagnosis

State "Diagnosing" is displayed.

Last operation time The latest time when a logic is executed is displayed.

(3) Logic list Information on logic execution is displayed.

Date The date and time when a log is output is displayed.

Unit name A unit name to be logged is displayed.

Logic name The name of a logic to be logged is displayed.

Log comment Logs related to data collection and logic execution are displayed.

For the displayed contents, refer to the following section.

Page 418 Log comment list

(1)

(3)

(2)

41822 Monitor22.1 Screen Configuration

Log comment listThe following table lists the log comments.

*1 It is also output when the logic 'Monitor' was deleted in the logic information list and then 'Switch to CSV reading mode' was performed in Reader.

Precautions

When a logic remaining value in the 'logic execution ended' log continues to increase, it indicates that the load limit is

exceeded. In this case, the number of logics to execute must be reduced.

Item Description

Logic registration*1 This log is output when a logic is started by Monitor.

Logic diagnostic start*1 This log is output when a logic is started by Monitor.

Unit data acquisition n (msec) m

section

This log is output when Reader generates section data and then Monitor read that section.

n: Reference for time required

m: Number of sections read together

Logic execution ended n (msec)

logics remaining (m)

This log is output when a logic execution ends.

n: Reference for execution time

m: Number of outstanding logics

Logic deletion This log is output when a logic is terminated by Monitor.

Logic execution error This log is output when an incomplete and unexecutable logic is executed (e.g., variable data was deleted by Reader

after the logic was created or the logic was deleted in 'Logic editing' in Analyzer).

This error does not occur after terminating the logic.

22 Monitor22.2 Monitor Setting 419

22

22.2 Monitor SettingThis section shows the settings for cooperating with an external tool such as Edgecross Basic Software.

Operating procedure

1. Click the [Setting] button.

2. Select [Logic setting] from the pull-down menu displayed in the upper portion of the screen.

Page 419 Logic setting


4. Select [Diagnosis result detail information file setting] from the pull-down menu displayed in the upper portion of the

screen.

Page 420 Diagnosis result detail information file setting


■ Logic setting

Window

Displayed items

*1 A wild car character (*) can be used. "*" represents one or more arbitrary characters.*2 '|' can be used as a separator when specifying multiple logic names.*3 Except for a case where logics are manually run from 'Logic editing' of Analyzer. Page 354 Manual execution*4 Do not specify a folder with the encryption attribute enabled.*5 Set it so that the path length of a file to be saved does not exceed 259 characters. The maximum path length of a pop up error log is as

follows:(Set folder)\ + 49 characters (if the logic name is 32 characters long)

Precautions

Monitor must be restarted to apply changes to the settings.

Item Description

The logic name to be automatically started is set to * (wild

card).*1*2Set a logic name (up to 256 characters) to automatically be added or started as a logic to be

diagnosed when starting Monitor.

Specify the folder that saves a pop up error log.*3*4*5 Set a save destination (up to 256 characters) of a log if the pop-up window appears when

executing a logic.

42022 Monitor22.2 Monitor Setting

■ Diagnosis result detail information file setting

Window

Displayed items

*1 Disabled if the checkbox of "The diagnosis result detail information file is output at error detection." is unselected. (The set value will be cleared.)

*2 Set it so that the path length of a file to be saved does not exceed 259 characters. For the name of a diagnosis result detail information file, refer to the following:Page 16 Real-time Statistic Diagnosis Tool

*3 Disabled if the checkbox of "The diagnosis result detail information file is output at error detection." is unselected. (The set value will be changed to '1000.')

*4 Only files that can be deleted are counted.*5 An error does not occur even if an file that cannot be deleted exists.

Precautions

• Monitor must be restarted to apply changes to the settings.

• A diagnosis result detail information file is not output if an output destination drive does not have space sufficient to output

the file. It will be output when the insufficiency is resolved. (Each file has approximately 300 B; therefore, to save 1000 files

which is the maximum, approximately 300 KB are required.)

Item Description

The diagnosis result detail information file is output at error

detection.

Select the checkbox to output a diagnosis result detail information file when an error is

detected.

Specify the folder in which the diagnosis result detail

information file is saved.*1*2Set the save destination (up to 256 characters) of a diagnosis result detail information file.

Specify the number of diagnosis result detail information files

to be held.(1-1000)*3*4*5Set the number of files to be retained.

If the number of output diagnosis result detail information files exceeds a specified number, files

are deleted in chronological order.

22 Monitor22.3 Starting Monitor 421

22

22.3 Starting MonitorThis section shows the operations for cooperating with Edgecross Basic Software.

Operating procedure


2. Check that "Reader_UNIT" is displayed.

"Reader_UNIT" is displayed when all of the following conditions are satisfied.






Monitor is restarted after the operation in is completed.

3. Logics are shown in the logic information list.

When "Diagnosing" is displayed in "State", cooperation with Edgecross Basic Software is available.

A logic indicated as "Monitor" in the "Logic" column is regarded as simple diagnosis (SPC, multivariate analysis (single central

value format) and guard band) in Analyzer.

4. When the required logic is not shown, right-click the logic information list and select the logic from [Logic start].

When Reader has read a CSV file, logics shown in the logic information list are diagnosed.

5. To stop the logic, right-click the logic and select [Logic stop].

Precautions

• The screen must remain in this state to keep Monitor running during diagnosis.

• When a logic set in an automatic start setting is stopped, it automatically restarts and its state shows "Diagnosing". To stop

the logic completely, perform either of the following operations:

- Stop the logic after deleting its logic name from an automatic start setting (Page 419 Monitor Setting)

- Delete the target logic in Analyzer (Page 255 Editing Logics)



If a function of Real-time Statistic Diagnosis Tool does not perform properly, check the applicable items in the following and

troubleshoot the error.


Unable to save, export, read, or

import of a file or a setting.

Does the path exceed 259

characters?

Set the path for save, export, reading, or import within 259 characters

including an extension.

Is the path of the access target an

external folder such as a shared

folder?

• Check whether it is accessible to the external folder.

• Check the following when cooperating with Real-time Flow Manager

installed in another industrial PC:

Is a cooperation destination folder accessible?

Is a communication cable connected?

Is a communication destination industrial PC running?

Is an external media recognized?

Are an access destination drive and media formatted in NTFS format?

Is there enough free space in a disk

specified as the save destination?

Secure free space in the disk.

When Reader reads a CSV file, the

following log is output and the file

cannot be read.

• "Because the time stamp that

overlapped with existing data

existed, data was not updated."

Check the time stamp of the data of

the CSV file*1 whether there is data

with a time stamp older than or same

as that of the latest data which was

already read.

Store data with a time stamp newer than that of the latest data which was

already read to the CSV input folder.

Unable to close Analyzer or Monitor. Is the [] button displayed on the

upper right of the screen in Analyzer

or Monitor?

• Right-click the icon of Analyzer or Monitor on the taskbar, and select [Close

window].


Press the + keys, or the Windows key + key to display the

application list.

Click the [] button on the upper right of the screen in Analyzer or Monitor

displayed in the list.

Unable to read a CSV file which is

stored in the specified folder.

Is Reader in the CSV file reading

mode?

Switch the operation mode of Reader to the CSV file reading mode.

Is Monitor running? (Can an

operation be performed in the screen

of Monitor?)

• Start Real-time Statistic Diagnosis Tool.

• Wait until the screen of Monitor is ready for operation.

A file is recognized as 'undefined' and

cannot be read even though its unit

definition exists.

Is a string of 13 characters set for

"Prefix of file name"?

Set a string from 1 to 12 characters or that of 14 or more characters for "Prefix

of file name."

An error (error code: 5812H) occurs

in a data diagnosis flow of Real-time

Flow Manager that cooperates with

Real-time Statistic Diagnosis Tool.

Does the diagnosis name have more

than 16 characters?

Set the diagnosis name to output to Real-time Flow Manager in the screen for

starting the diagnosis.

A dialog displayed in Analyzer is

hidden behind an inactive screen and

cannot be operated.

Is a dialog with the icon of Analyzer in

the application list that is displayed by

pressing the + keys or the

Windows key + key?

Perform the following procedure.

Press the + keys, or the Windows key + key to display the

application list.

Click the [] button on the upper right of the dialog in Analyzer displayed in

the list.

Close the inactive screens other than Analyzer, or move the screens so

that they do not overlap with the dialog.

A screen becomes inactive in

Analyzer and cannot be operated.

Contents in the screen are not

displayed properly.

Is a value other than 100% set for

"Change the size of text, apps, and

other items" in Windows?

• Change the value to 100% for "Change the size of text, apps, and other

items".

• For Windows 10 (version 1703 or later)*2, the display of a screen can be

enlarged with high DPI scaling by using a Windows 10 function*3.

Select the execution file*4,*5, then select [Properties] on the right-click

menu.

Select the checkbox of "Override high DPI scaling behavior. Scaling

performed by:" in the [Compatibility] tab, then select "System" from the pull-

down list.

Click the [OK] button.

Read data is not displayed though

external media is set for 'Reader

equipment data folder' and 'RSD

dataset folder'.

Is the external media formatted in

other than NTSF format?

For the checking method, refer to the

following:


Format the external media in NTFS format.


23

*1 The CSV file is stored in a folder under the "Reader equipment data folder".Reader_UNIT\YYMMDD\Original\Invalid('YYMMDD' indicates the current date.)

*2 The Windows version can be checked by the following procedure. Press Windows key + , or select [Windows System] [Run] from Windows Start. Enter 'winver' in the "Run" screen. Check the version in the displayed screen.

*3 The display will be blurred by enlarging.The following lists the setting values for "Change the size of text, apps, and other items" and the recommended display resolution for each setting value in Windows 10.Setting value: 100%, display resolution: 1280 800 dots or moreSetting value: 125%, display resolution: 2048 1536 dots or moreSetting value: 150%, display resolution: 2048 1536 dots or moreSetting value: 175%, display resolution: 2880 1620 dots or moreSetting value: 200%, display resolution: 2880 1620 dots or moreSetting value: 225%, display resolution: 3840 2160 dots or moreSetting value: 250%, display resolution: 3840 2160 dots or more

*4 The execution file is stored in the folder in which Real-time Data Analyzer is installed.The following are examples of storage locations.(Example) When the installation destination folder is set as the default:Screen of Reader (setting edit mode): C:\Mitsubishi Electric Edgecross Software\RDA\RSD\en\Reader\Reader.exeScreen of Reader (CSV file reading mode): C:\Mitsubishi Electric Edgecross Software\RDA\RSD\en\Reader\Reader_UNIT.exeScreen of Analyzer: C:\Mitsubishi Electric Edgecross Software\RDA\RSD\en\Analyzer\Analyzer.exeScreen of Monitor: C:\Mitsubishi Electric Edgecross Software\RDA\RSD\en\Monitor\Monitor.exe

*5 The following are examples of storage locations for the execution file selected when enlarging the display of a screen of License Manager and the utility for switching language.(Example) When the installation destination folder is set as the default:License Manager: C:\Mitsubishi Electric Edgecross Software\License Manager\License Manager.exeUtility for switching language: C:\Mitsubishi Electric Edgecross Software\RDA\LanguageSwitchingUtility\RDA_LanguageSwitchingUtility.exe

424APPXAppendix 1 Processing Performance and Processing Time

APPENDIXAppendix 1 Processing Performance and Processing

TimeThis section shows the processing performance and processing time of Real-time Data Analyzer.

Similar Waveform Recognition ToolThe following shows the processing performance and processing time of Similar Waveform Recognition Tool.

Measurement conditionsThe following industrial PCs are used for measurement.

■ Industrial PC (operation-guaranteed environment)

■ Industrial PC (MELIPC MI5122-VW)

For MELIPC MI5122-VW, refer to the following manual.

MELIPC MI5000 Series User's Manual (Startup)

■ Industrial PC (MELIPC MI1002-W)

These industrial PC are used as installation destinations for Real-time Flow Manager when cooperating with the Real-time

Flow Manager installed in another industrial PC.

For MELIPC MI1002-W, refer to the following:

MI1002-W User's Manual

Item Description

CPU Intel Core i3 (1.90GHz Quad Core)

RAM 8 GB

Operating system

(64-bit version)

Windows 10 IoT Enterprise 2016 LTSB

Ethernet network I/F Interface 1000BASE-T

Communication method Full-duplex

Communication speed 1000Mbps (1000BASE-T)

Item Description

CPU Intel Core i7 (2.6GHz Quad Core)

RAM 12 GB

Operating system

(64-bit version)





Item Description

CPU Intel Atom E3826 (1.46GHz Dual Core)

RAM 4GB

Operating system

(64-bit version)





APPXAppendix 1 Processing Performance and Processing Time 425

A

Processing performance (waveform learning)The following shows the processing time that takes to learn a waveform.

• The time taken for Similar Waveform Recognition Tool to learn a waveform is greatly affected not only by the

number of samples or the unit waveform width but also by the waveform properties such as cycle stability,

S/N ratio, instability. The performance value will be the average of the measurement time when five types of

waveforms with different properties are processed.

The processing time may significantly be longer when processing a waveform that is not recommended as

a reference waveform because the cycle is unclear.

• Waveform learning processing can be performed for up to three days, and the operation is not guaranteed

for learning exceeding three days.

• The processing time mentioned above indicates a time per learning. When performing it two or more times,

the above processing time is repeated for the number of learning settings. (Learning can be performed up to

64 times in one setting in Similar Waveform Recognition Tool.)

■ Industrial PC (operation-guaranteed environment): Not applicable because the calculation time exceeds three days.

: Not applicable because the number of samples of a reference waveform exceeds the maximum number of samples of

learning (1000000).

(Unit: minute)

*1 If the length of the reference waveform is 'unit waveform width 800', the number of samples of the reference waveform is 160000 when the unit waveform width is 200.

■ Industrial PC (MELIPC MI5122-VW): Not applicable because the calculation time exceeds three days.

: Not applicable because the number of samples of a reference waveform exceeds the maximum number of samples of

learning (1000000).

(Unit: minute)

Length of a reference waveform*1

Unit waveform width (record)

200 400 600 800 1000

Unit waveform width 200 3 28 101 299 734

Unit waveform width 400 14 117 382 985 3977

Unit waveform width 600 33 239 1023 3281

Unit waveform width 800 60 354 1855

Unit waveform width 1000 92 719 2944

Unit waveform width 2000 249 3020

Unit waveform width 3000 942





200 400 600 800 1000

Unit waveform width

200

2 13 47 145 357

Unit waveform width

400

7 56 178 469 1092

Unit waveform width

600

16 114 376 1697

Unit waveform width

800

30 169 960

Unit waveform width

1000

46 261 1523

Unit waveform width

2000

129 1562


*1 If the length of the reference waveform is 'unit waveform width 800', the number of samples of the reference waveform is 160000 when the unit waveform width is 200.

Unit waveform width

3000

271

Unit waveform width

4000

884

Unit waveform width

5000

1402



200 400 600 800 1000


A

Processing performance (similarity diagnosis)The following shows the maximum number of diagnoses that can be set and the CPU usage rate.

• The numbers in the above tables are the number of diagnoses that can be set when the number of data

diagnosis flow of Real-time Flow Manager is '1'. When setting similarity diagnoses for two or more data

diagnosis flows separately, the number of diagnoses exceeding the upper limit of the numbers in the above

tables can be set.

• Reference waveform learning data created when performing waveform learning with the length of the

reference waveform set to 'unit waveform width 200' is used.

• The set diagnosis cycle is '5'.

• CPU usage rate when the upper limit is set depends on the number of CPU cores and the number of data

diagnosis flows. The maximum CPU usage rate of Similar Waveform Recognition Tool is approximately as

follows:

100/number of CPU cores number of diagnosis flows

• A value of the CPU usage rate is the average value of the CPU usage rate for 10 minutes after the

diagnosis status turns into the 'similar' state or 'different' state from the 'preparing for diagnosis' state.

■ Using one industrial PC • Industrial PC (operation-guaranteed environment)

• Industrial PC (MELIPC MI5122-VW)

: The data collector is not available under the given conditions.

: No description since the data collector is unavailable under the given conditions.

Collection cycle (ms)

Data collector to use Collection method

Measurement target


200 400 600 800 1000

100 SLMP Data Collector

(MELSEC connection)

Number of

diagnoses

64 16 8 2 2

CPU usage rate

(%)

49 51 55 48 58



Measurement target


200 400 600 800 1000

0.5 CC-Link IE TSN Data

Collector

Cycle setting Number of

diagnoses

4

CPU usage rate

(%)

60

1 CC-Link IE Field Network

Data Collector

Each link scan Number of

diagnoses

4

CPU usage rate

(%)

36


Data Collector


diagnoses

4

CPU usage rate

(%)

88


Data Collector

Cycle

specification

Number of

diagnoses

8 1

CPU usage rate

(%)

71 53


Data Collector

Cycle

specification

Number of

diagnoses

16 4 1

CPU usage rate

(%)

96 59 43


For details on a data collector to use, refer to the following manuals.

SLMP Data Collector User's Manual

CC-Link IE Field Network Data Collector (MELIPC MI5122-VW) User's Manual


Data Collector

Cycle

specification

Number of

diagnoses

64 16 4 2 2

CPU usage rate

(%)

86 54 26 19 28


(MELSEC connection)

Number of

diagnoses

64 16 8 4 4

CPU usage rate

(%)

27 96 62 39 37



Measurement target


200 400 600 800 1000


A

■ Using two separate industrial PCs • Installation destination for Similar Waveform Recognition Tool: Page 424 Industrial PC (operation-guaranteed

environment)

Installation destination for Real-time Flow Manager: Page 424 Industrial PC (MELIPC MI1002-W)

• Installation destination for Similar Waveform Recognition Tool: Page 424 Industrial PC (operation-guaranteed

environment)

Installation destination for Real-time Flow Manager: Page 424 Industrial PC (operation-guaranteed environment)

• Installation destination for Similar Waveform Recognition Tool: Page 424 Industrial PC (operation-guaranteed

environment)

Installation destination for Real-time Flow Manager: Page 424 Industrial PC (MELIPC MI5122-VW)

• Installation destination for Similar Waveform Recognition Tool: Page 424 Industrial PC (MELIPC MI5122-VW)

Installation destination for Real-time Flow Manager: Page 424 Industrial PC (MELIPC MI1002-W)


Installation destination for Real-time Flow Manager: Page 424 Industrial PC (operation-guaranteed environment)



Measurement target


200 400 600 800 1000


(MELSEC connection)

Number of

diagnoses

64 16 8 2 2

CPU usage rate

(%)

17 16 16 15 16



Measurement target


200 400 600 800 1000


(MELSEC connection)

Number of

diagnoses

64 16 8 2 2

CPU usage rate

(%)

20 14 10 10 11



Measurement target


200 400 600 800 1000


(MELSEC connection)

Number of

diagnoses

64 16 8 2 2

CPU usage rate

(%)

25 13 10 18 10



Measurement target


200 400 600 800 1000


(MELSEC connection)

Number of

diagnoses

64 16 8 2 2

CPU usage rate

(%)

25 17 15 12 13



Measurement target


200 400 600 800 1000


(MELSEC connection)

Number of

diagnoses

64 16 8 2 2

CPU usage rate

(%)

25 50 28 16 21



Installation destination for Real-time Flow Manager: Page 424 Industrial PC (MELIPC MI5122-VW)

: The data collector is not available under the given conditions.

: No description since the data collector is unavailable under the given conditions.

For details on a data collector to use, refer to the following manuals.

SLMP Data Collector User's Manual

CC-Link IE Field Network Data Collector (MELIPC MI5122-VW) User's Manual



Measurement target


200 400 600 800 1000


Data Collector


diagnoses

4

CPU usage rate

(%)

20


Data Collector


diagnoses

4

CPU usage rate

(%)

13


Data Collector

Cycle

specification

Number of

diagnoses

8 1

CPU usage rate

(%)

13 10


Data Collector

Cycle

specification

Number of

diagnoses

16 4 1

CPU usage rate

(%)

25 10 10


Data Collector

Cycle

specification

Number of

diagnoses

64 16 4 2 2

CPU usage rate

(%)

40 18 13 10 10


(MELSEC connection)

Number of

diagnoses

64 16 8 4 4

CPU usage rate

(%)

18 10 18 13 13


A

CPU usage rateIt is recommended to limit the CPU usage rate of the whole industrial PC to average 60% so that Similar Waveform

Recognition Tool can be operated stably. If the CPU average usage rate exceeds 60%, review the settings and system

configuration by referring this measurement result to reduce the load on the industrial PC.

■ Example for reducing the load on an industrial PC by modifying the settings • Before modification

• After modification (changing the unit waveform width and the number of diagnoses)

■ Example for reducing the load on an industrial PC by modifying the system configuration • Before modification

• After modification (changing the installation destination for Real-time Flow Manager)

Installation destination for Similar Waveform Recognition Tool

Installation destination for Real-time Flow Manager


Number of diagnoses Collection cycle (ms)

Page 424 Industrial PC

(MELIPC MI5122-VW)

Same industrial PC as the

installation destination for

Similar Waveform

Recognition Tool

400 16 100

(SLMP Data Collector

(MELSEC connection))






(MELIPC MI5122-VW)



Similar Waveform

Recognition Tool

600 8 100








(MELIPC MI5122-VW)



Similar Waveform

Recognition Tool

400 16 100








(MELIPC MI5122-VW)


(MELIPC MI1002-W)

400 16 100




Real-time Statistic Diagnosis ToolThe following shows the processing performance and processing time of Real-time Statistic Diagnosis Tool.

Measurement conditions

■ Industrial PCRefer to the following section.

Page 424 Industrial PC (MELIPC MI5122-VW)

■ Parameter setting value

Processing performance

■ Diagnosis processing (per diagnosis)(Unit: second)

Setting item Setting content

CSV file Collection cycle 1000 ms

Number of rows 10 to 36000 rows

Number of variables 6 (excluding the time stamp column)

Variable type Numeric

Diagnosis contents The following logics to diagnose MT distance:

Multi trace statistic (statistic:range)

MVA statistic (processing type: MT distance, target: 6 variables)

Limit pair diagnostics (target: MT distance calculated in )

Write log and error notification

Number of rows in a CSV file Performance value

10 5.0

100 5.0

1000 5.0

10000 5.0

30000 5.0

36000 5.0

APPXAppendix 2 Available Characters 433

A

Appendix 2 Available CharactersThis section shows the characters that can be used in Real-Time Data Analyzer.

Similar Waveform Recognition ToolThe following table shows the characters that can be used in Similar Waveform Recognition Tool.

Screen Item Available/invalid character Reference

Edgecross Link Setting Port number ■Available characters

• Single-byte characters (decimal)

Page 35

Edgecross

Link SettingUser name ■Available characters

• Printable ASCII characters (0x20 to

0x7E)

Password ■Available characters

• Printable ASCII characters (0x20 to

0x7E)

Reference Waveform Learning

Data File Creation setting

Folder to save the Reference Waveform File ■Available characters

• Characters available for file paths

in Windows

Page 63

Reference

Waveform

Learning Data

File Creation

Setting Screen

Reference waveform ■Available characters


Reference waveform learning

data output file path

Output folder ■Available characters


in Windows

File Name ■Available characters


in Windows

However, the reserved words

described in the following section

cannot be used.

Reserved words that

cannot be used

Setting of reference waveform

learning data creation

Learning period ■Available characters


Unit waveform width ■Available characters


Publishing Data Setting Setting name ■Available characters

• Characters available for file names

in Windows

Page 66

Publishing

Data Setting

ScreenDiagnosis execution cycle ■Available characters


434APPXAppendix 2 Available Characters

Screen Item Available/invalid character Reference

Operation Settings Setting name ■Invalid characters

• Space

• ASCII control code (U+0000 to

U+001F), DEL(U+007F)

Page 69

Operation

Setting Screen

Reference waveform learning data file ■Available characters


in Windows



cannot be used.

Page 435 Reserved words that

cannot be used

I/O setting Response Data Name ■Invalid characters

• , (comma), space

• ASCII control code (U+0000 to

U+001F), DEL(U+007F)

• Unicode (U+00A0, U+00AD,

U+2028, U+2029)

Diagnosis operation setting Detection sensitivity setting

(when entering a value directly)

■Available characters


• . (period)

Diagnosis result file saving

setting

Number of digits after decimal

point



Switched when the following

number of rows (records) is

exceeded.



Switching cycle setting ■Available characters


Max. number of save files

(when exceeded, the oldest file

is deleted.)



Save destination folder ■Available characters


in Windows

Prefix of file name ■Available characters


in Windows



cannot be used.

Page 435 Reserved words that

cannot be used

Diagnosis stop condition

setting

Constant value For the following data types:

• INT

• UINT

• DINT

• UDINT


• Numerals (decimal)

• +, -

Page 74

Diagnosis Stop

Condition

Setting Screen

For the following data types:

• REAL

• LREAL



• e, E

• +, -

• . (period)

For the STRING type ■Available characters

Printable ASCII characters (0x20 to

0x7E)

For the WSTRING type ■Invalid characters

Unicode(U+0000 to U+001F, U+007F

to U+00A0, U+00AD, U+2028,

U+2029)


A

Reserved words that cannot be used

Reserved word

CON

AUX

COM0

COM1

COM2

COM3

COM4

COM5

COM6

COM7

COM8

COM9

LPT0

LPT1

LPT2

LPT3

LPT4

LPT5

LPT6

LPT7

LPT8

LPT9

PRN

NUL


Real-time Statistic Diagnosis ToolThe following table shows the characters that can be used in Real-time Statistic Diagnosis Tool.

For the characters that can be used when setting parameters of each primitive for logic editing, refer to the following section.

Primitive parameters

Operation Item Available/invalid character Reference

Folder definition Reader equipment data folder ■Invalid characters

• ", *, /, <, >, ?, |

Page 129 Folder

DefinitionCSV input folder

RSD dataset folder

Adding a unit definition Sample CSV file ■Invalid characters

• ", *, /, <, >, ?, |

Page 131 Adding a Unit

Definition

Reading target CSV file name ■Invalid characters

• ", /, :, <, >, ?, \, |

Unit name ■Invalid characters

• Space

• Single-byte katakana

• Double-byte character

• !, ", #, $, %, &, ', (, ), *, +, ,

(comma), ., /, :, ;, <, =, >, ?, @,

[, \, ], ^, _, `, {, |, }, ~


described in the following

section cannot be used.

Page 439 Reserved words

that cannot be used

Editing settings Adding variables Sample CSV file ■Invalid characters

• ", *, /, <, >, ?, |

Page 148 Adding

variables

Reading target CSV file name ■Invalid characters

• ", /, :, <, >, ?, \, |

Defining an extended

section

Extended section name ■Invalid characters

• Space



• !, ", #, $, %, &, ', (, ), *, +, ,

(comma), ., /, :, ;, <, =, >, ?, @,

[, \, ], ^, _, `, {, |, }, ~





that cannot be used

Page 155 Defining an

extended section

Editing an SPC specification value Upper limit ■Available characters


• e, E

• +, -

• . (period)

Page 199 Editing an

SPC specification valueTarget

Lower limit

Sigma

Creating a correlation matrix (creating a whole

matrix)

Matrix storage folder ■Invalid characters

• ", *, /, <, >, ?, |





that cannot be used

Page 225 Creating

whole matrix

Reading folder ■Invalid characters

• ", *, /, <, >, ?, |

Please input Matrix name first. ■Invalid characters

• ", *, /, :, <, >, ?, \, |





that cannot be used

Variable matrix name


A

*1 The following symbols cannot be entered.", *, /, :, <, >, ?, \, |

Creating a correlation matrix (creating a partial

matrix)

Partial matrix name (It is possible to save

it also with Enter key).

■Invalid characters

• ", *, /, :, <, >, ?, \, |




Reserved words

that cannot be used

Page 231 Creating

partial matrix

Logic editing (creating a logic) Logic name ■Invalid characters*1

• ", *, /, :, <, >, ?, \, |





that cannot be used

Page 258 Creating logic

Output ■Invalid characters*1

• ", *, /, :, <, >, ?, \, |

Logic variable setting Variable name ■Invalid characters*1

• ", *, /, :, <, >, ?, \, |

Page 357 Screen

configuration

Logic variable setting

(selecting and setting a

variable type)

Variable name ■Invalid characters*1

• ", *, /, :, <, >, ?, \, |

Page 358 Selecting and

setting variable type

When setting an

execution conversion

table

Upper limit ■Available characters

• -, . (period)

• 0 to 9

Page 360 Setting up

execution conversion

table

Lower limit ■Available characters

• -, . (period)

• 0 to 9

Logic setting Specify the folder that saves a pop up

error log.


• ", *, /, <, >, ?, |





that cannot be used

Page 393 Logic Setting

Monitor setting Logic setting Specify the folder that saves a pop up

error log.


• ", *, /, <, >, ?, |





that cannot be used

Page 419 Logic setting

The logic name to be automatically

started is set to * (wild card).


• ", /, :, <, >, ?, \

Diagnosis result detail

information file setting

Specify the folder in which the diagnosis

result detail information file is saved.


• ", *, /, <, >, ?, |





that cannot be used

Page 420 Diagnosis

result detail information

file setting

Operation Item Available/invalid character Reference


Primitive parametersThe following table shows the characters that can be used when setting parameters of each primitive for logic editing.

Primitive Item Available/invalid character Reference

Waveform conversion (when selecting

"Wave statistic operation" for

"Conversion type")

Expression ■Invalid characters

• Single-byte alphabetical characters

(upper-case characters)



• !, ", #, $, %, &, ', , (comma), ., :, ;, <, =, >,

?, @, [, \, ], _, `, {, |, }, ~

Setting items when

selecting "Wave statistic

operation" for

"Conversion type"

Multi-section calculation Expression ■Invalid characters





• !, ", #, $, %, &, ', , (comma), ., :, ;, <, =, >,

?, @, [, \, ], _, `, {, |, }, ~

Page 313 Multi-section

calculation

Calculation in present section Expression ■Invalid characters





• !, ", #, $, %, &, ', , (comma), ., :, ;, <, =, >,

?, @, [, \, ], _, `, {, |, }, ~

Page 314 Calculation in

present section

Regression operation Matrix including regression equation ■Invalid characters

• ", *, /, <, >, ?, |

Page 317 Regression

operation

MVA statistic Criterion matrix ■Invalid characters

• ", *, /, <, >, ?, |

Page 318 MVA statistic

Save CSV data Specify CSV output file name. ■Invalid characters

• ", *, /, <, >, ?, |

Page 340 Save CSV

data

Write template Output folder ■Invalid characters

• ", *, /, <, >, ?, |

Page 345 Write

template

File name ■Invalid characters

• ", *, /, :, <, >, ?, \, |

Parametrical conversion Conversion table ■Invalid characters

• ", *, /, <, >, ?, |

Page 347 Parametrical

conversion

Read measurement data. Specify data folder ■Invalid characters

• ", *, /, <, >, ?, |

Page 351 Read

measurement data


A

Reserved words that cannot be used

Reserved word

CON

AUX

COM0

COM1

COM2

COM3

COM4

COM5

COM6

COM7

COM8

COM9

LPT0

LPT1

LPT2

LPT3

LPT4

LPT5

LPT6

LPT7

LPT8

LPT9

PRN

NUL

440APPXAppendix 3 Environment Settings for an Industrial PC where Real-time Data Analyzer Runs

Appendix 3 Environment Settings for an Industrial PC where Real-time Data Analyzer Runs

This section shows the environment settings for an industrial PC where Real-time Data Analyzer runs.

Settings for indexes created for Windows SearchThe load due to indexes created for Windows Search which is one of the standard features of Windows 10 may affect

operations of Real-time Data Analyzer, therefore, change the setting as necessary.

Operating procedure

1. Open the control panel and change the setting of "View by" to "Small Icons".

2. Select "Indexing Options".

3. Click the [Modify] button.

4. Select a location in "Summary of selected locations", and unselect the checkbox of a folder where index does not need to

be created in "Change selected locations".

5. Repeat step 4 for all locations displayed in "Summary of selected locations".


APPXAppendix 3 Environment Settings for an Industrial PC where Real-time Data Analyzer Runs 441

A

7. Click the [Advanced] button.

8. Click the [Rebuild] button.

9. Click the [OK] button in the "Rebuild Index" screen.

10.Click the [Close] button.

442APPXAppendix 3 Environment Settings for an Industrial PC where Real-time Data Analyzer Runs

11. To exclude the contents of a file from a search, right-click the C drive in Windows Explorer and select [Properties].

12.Unselect the checkbox of "Allow files on this drive to have contents indexed in addition to file properties", and click the

[OK] button.

13.Click the [OK] button in the "Confirm Attribute Changes" screen.

APPXAppendix 3 Environment Settings for an Industrial PC where Real-time Data Analyzer Runs 443

A

Settings for the Customer Experience Improvement ProgramThe Customer Experience Improvement Program which is one of the standard features of Windows 10 may affect operations

of Real-time Data Analyzer, therefore, disable the setting as necessary.

Operating procedure

1. Change the setting of "Windows should ask for my feedback" to "Never".

[Settings] [Privacy] [Feedback & diagnostics] from Windows Start.

2. Restart the industrial PC.

444APPXAppendix 4 System Configuration and Application Settings for Operation of Real-time Data Analyzer

Appendix 4 System Configuration and Application Settings for Operation of Real-time Data Analyzer

When operating Real-time Data Analyzer with other Edge applications, the load on an industrial PC may increase depending

on the settings; an error or data corruption may occur in each application.

It is recommended to set each application to limit the CPU and memory usage rate of the whole industrial PC to average

60%*1 so that each application can be operated stably.

This section shows the examples of system configurations and application settings to stabilize operation.

*1 If the usage rate exceeds 60%, it is recommended to cooperate with Real-time Flow Manager that is installed in another industrial PC.

System configurationDiagnose and monitor the operation status of one production line that consists of ten devices simultaneously in a MELIPC.

Item Content

Industrial PC MELIPC MI5122-VW

Access target device QCPU10 (via QJ71E71-100)

Data collector SLMP Data Collector

Edge application Real-time Data Analyzer

MC Works64 EdgeComputing Edition

APPXAppendix 4 System Configuration and Application Settings for Operation of Real-time Data Analyzer 445

A

Application settingsPurpose Application Description

Diagnostic • Real-time Data Analyzer

• Edgecross Basic Software

(Data diagnosis flow)

• GX LogViewer

■Data diagnosis flow setting No.1

Collection interval 500 milliseconds

Number of collection points 100

Diagnostic method Real-time Statistic Diagnosis Tool (MT method)

Setting or not setting for an

extended section

Not set

Reading target variable 100 variables

Statistic extraction Multi trace statistic

(Calculate from 6 variables in each

section)

Criterion matrix size 6 columns105 rows

Number of pieces of feedback 1

■Data diagnosis flow setting No.2



Diagnostic method Similar Waveform Recognition Tool

Data type of diagnosis target

data

REAL

Number of diagnoses 1

Monitoring or not monitoring

the result of a similar

waveform recognition in GX

LogViewer

Not set

Number of pieces of feedback 1

Monitor • MC Works64 EdgeComputing

Edition

• Edgecross Basic Software

(Management Shell)

Number of monitoring items 1400

Details

Data display (GraphWorX) 200

Alarm monitoring (AlarmWorX

Server)

300

Logging (MC Historian) 900 (Data collection rate: 1

second)

Number of display screens 1

Screen display and setting content

Data display (Process Point) 200

Alarm list display

(AlarmWorX64 Viewer)

Real Time Alarm display

Animation Speed

(1000 milliseconds)

Graph display

(TrendWorX64 Viewer)

four line graphs (using logging

data)

Trend Range and Period: 5

minutes

Animation Speed: 1 second

History Read Rate: 5 seconds

Screen Update Interval (Scan

Rate)

1000 milliseconds

3D View Not set

Data storing • Edgecross Basic Software (data

logging flow)

■Data logging flow setting No.1 to No.2



Data storing format File

446APPXAppendix 4 System Configuration and Application Settings for Operation of Real-time Data Analyzer

Corrective actions for high load on an industrial PCWhen the load on an industrial PC is high, perform the following operations:

• Refer to 'The load on an industrial PC is high.' described in the following section, and take the corrective action.

(Page 98 TROUBLESHOOTING BY SYMPTOM)

• Refer to 'TROUBLESHOOTING' in the following manual, and reduce the load.

(MC Works64 Version 4 Startup Manual)

APPXAppendix 5 Software Licenses 447

A

Appendix 5 Software LicensesThis software consists of multiple software components. Each of them is copyrighted by Mitsubishi and/or third parties.

The following types of software are contained:

● Software copyrighted by third parties and distributed as free software

● Software licensed under the GNU Lesser General Public License (LGPL)

● Software licensed under the Eclipse Public License (EPL) and Eclipse Distribution License (EDL)

For the open source software of the LGPL and EPL used in this software, it is allowed to obtain and redistribute the source

codes according to the LGPL and EPL conditions described in the following software information.

Page 447 Software information

These source codes can be downloaded from the Edgecross Consortium website (www.edgecross.org).

These are distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY, without even the implied warranty of

MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. In addition, source codes are not distributed for the

software that Mitsubishi and/or third parties have copyrights to.

Please refrain from inquiring about the source codes of this open source.

Software informationThis product contains the following software:

RapidJSON (Page 447 RapidJSON)

msinttypes (Page 448 msinttypes)

mosquitto ( mosquitto)

OpenSSL toolkit (Page 453 OpenSSL toolkit)

LabVIEW Runtime, LabWindows/CVI Runtime (Page 456 LabVIEW Runtime, LabWindows/CVI Runtime)

RapidJSONThis product uses RapidJSON licensed under the MIT License.

The copyright and permission notices of RapidJSON are described below.Tencent is pleased to support the open source community by making RapidJSON available.

Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. All rights reserved.

If you have downloaded a copy of the RapidJSON binary from Tencent, please note that the RapidJSON binary is licensed under the MIT License.

If you have downloaded a copy of the RapidJSON source code from Tencent, please note that RapidJSON source code is licensed under the MIT License,

except for the third-party components listed below which are subject to different license terms. Your integration of RapidJSON into your own projects may

require compliance with the MIT License, as well as the other licenses applicable to the third-party components included within RapidJSON. To avoid the

problematic JSON license in your own projects, it's sufficient to exclude the bin/jsonchecker/ directory, as it's the only code under the JSON license.

A copy of the MIT License is included in this file.

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the

Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software,

and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE

WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR

COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR

OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

448APPXAppendix 5 Software Licenses

msinttypesThis product uses msinttypes licensed under the BSD License.

The copyright and permission notices of msinttypes are described below.The msinttypes r29

Copyright (c) 2006-2013 Alexander Chemeris

All rights reserved.

Redistribution and use in source and binary forms, with or without

modification, are permitted provided that the following conditions are met:

* Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.

* Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other

materials provided with the distribution.

* Neither the name of copyright holder nor the names of its contributors may be used to endorse or promote products derived from this software without specific

prior written permission.

THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT

NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT

SHALL THE REGENTS AND CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL

DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR

BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT

(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY

OF SUCH DAMAGE.


A

mosquittoThis product uses mosquitto licensed under the Eclipse Public License - v 1.0 and the Eclipse Distribution License - v 1.0.

The permission notice of the Eclipse Public License - v 1.0 and Eclipse Distribution License - v1.0 is described below.Eclipse Public License - v 1.0

THE ACCOMPANYING PROGRAM IS PROVIDED UNDER THE TERMS OF THIS ECLIPSE PUBLIC LICENSE ("AGREEMENT"). ANY USE,

REPRODUCTION OR DISTRIBUTION OF THE PROGRAM CONSTITUTES RECIPIENT'S ACCEPTANCE OF THIS AGREEMENT.

1. DEFINITIONS

"Contribution" means:

a) in the case of the initial Contributor, the initial code and

documentation distributed under this Agreement, and

b) in the case of each subsequent Contributor:

i) changes to the Program, and

ii) additions to the Program;

where such changes and/or additions to the Program originate from and are

distributed by that particular Contributor. A Contribution 'originates' from a

Contributor if it was added to the Program by such Contributor itself or anyone

acting on such Contributor's behalf. Contributions do not include additions to

the Program which: (i) are separate modules of software distributed in conjunction with the Program under their own license agreement, and (ii) are

not derivative works of the Program.

"Contributor" means any person or entity that distributes the Program.

"Licensed Patents " mean patent claims licensable by a Contributor which are

necessarily infringed by the use or sale of its Contribution alone or when combined with the Program.

"Program" means the Contributions distributed in accordance with this Agreement.

"Recipient" means anyone who receives the Program under this Agreement, including all Contributors.

2. GRANT OF RIGHTS

a) Subject to the terms of this Agreement, each Contributor hereby grants

Recipient a non-exclusive, worldwide, royalty-free copyright license to

reproduce, prepare derivative works of, publicly display, publicly

perform, distribute and sublicense the Contribution of such Contributor,

if any, and such derivative works, in source code and object code form.

b) Subject to the terms of this Agreement, each Contributor hereby grants

Recipient a non-exclusive, worldwide, royalty-free patent license under

Licensed Patents to make, use, sell, offer to sell, import and otherwise

transfer the Contribution of such Contributor, if any, in source code and

object code form. This patent license shall apply to the combination of the

Contribution and the Program if, at the time the Contribution is added by the

Contributor, such addition of the Contribution causes such combination to be

covered by the Licensed Patents. The patent license shall not apply to any

other combinations which include the Contribution. No hardware per se is

licensed hereunder.

c) Recipient understands that although each Contributor grants the licenses

to its Contributions set forth herein, no assurances are provided by any

Contributor that the Program does not infringe the patent or other

intellectual property rights of any other entity. Each Contributor disclaims

any liability to Recipient for claims brought by any other entity based on

infringement of intellectual property rights or otherwise. As a condition to

exercising the rights and licenses granted hereunder, each Recipient hereby

assumes sole responsibility to secure any other intellectual property rights

needed, if any. For example, if a third party patent license is required to


allow Recipient to distribute the Program, it is Recipient's responsibility

to acquire that license before distributing the Program.

d) Each Contributor represents that to its knowledge it has sufficient

copyright rights in its Contribution, if any, to grant the copyright license

set forth in this Agreement.

3. REQUIREMENTS

A Contributor may choose to distribute the Program in object code form under

its own license agreement, provided that:

a) it complies with the terms and conditions of this Agreement; and

b) its license agreement:

i) effectively disclaims on behalf of all Contributors all warranties and

conditions, express and implied, including warranties or conditions of

title and non-infringement, and implied warranties or conditions of

merchantability and fitness for a particular purpose;

ii) effectively excludes on behalf of all Contributors all liability for

damages, including direct, indirect, special, incidental and consequential

damages, such as lost profits;

iii) states that any provisions which differ from this Agreement are offered

by that Contributor alone and not by any other party; and

iv) states that source code for the Program is available from such

Contributor, and informs licensees how to obtain it in a reasonable manner

on or through a medium customarily used for software exchange.

When the Program is made available in source code form:

a) it must be made available under this Agreement; and

b) a copy of this Agreement must be included with each copy of the Program.

Contributors may not remove or alter any copyright notices contained within

the Program.

Each Contributor must identify itself as the originator of its Contribution,

if any, in a manner that reasonably allows subsequent Recipients to identify

the originator of the Contribution.

4. COMMERCIAL DISTRIBUTION

Commercial distributors of software may accept certain responsibilities with

respect to end users, business partners and the like. While this license is

intended to facilitate the commercial use of the Program, the Contributor who

includes the Program in a commercial product offering should do so in a

manner which does not create potential liability for other Contributors.

Therefore, if a Contributor includes the Program in a commercial product

offering, such Contributor ("Commercial Contributor") hereby agrees to defend

and indemnify every other Contributor ("Indemnified Contributor") against any

losses, damages and costs (collectively "Losses") arising from claims,

lawsuits and other legal actions brought by a third party against the

Indemnified Contributor to the extent caused by the acts or omissions of such

Commercial Contributor in connection with its distribution of the Program in

a commercial product offering. The obligations in this section do not apply

to any claims or Losses relating to any actual or alleged intellectual


A

property infringement. In order to qualify, an Indemnified Contributor must:

a) promptly notify the Commercial Contributor in writing of such claim, and

b) allow the Commercial Contributor to control, and cooperate with the

Commercial Contributor in, the defense and any related settlement

negotiations. The Indemnified Contributor may participate in any such claim

at its own expense.

For example, a Contributor might include the Program in a commercial product

offering, Product X. That Contributor is then a Commercial Contributor. If

that Commercial Contributor then makes performance claims, or offers

warranties related to Product X, those performance claims and warranties are

such Commercial Contributor's responsibility alone. Under this section, the

Commercial Contributor would have to defend claims against the other

Contributors related to those performance claims and warranties, and if a

court requires any other Contributor to pay any damages as a result, the

Commercial Contributor must pay those damages.

5. NO WARRANTY

EXCEPT AS EXPRESSLY SET FORTH IN THIS AGREEMENT, THE PROGRAM IS PROVIDED ON

AN "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, EITHER

EXPRESS OR IMPLIED INCLUDING, WITHOUT LIMITATION, ANY WARRANTIES OR

CONDITIONS OF TITLE, NON-INFRINGEMENT, MERCHANTABILITY OR FITNESS FOR A

PARTICULAR PURPOSE. Each Recipient is solely responsible for determining the

appropriateness of using and distributing the Program and assumes all risks

associated with its exercise of rights under this Agreement , including but

not limited to the risks and costs of program errors, compliance with

applicable laws, damage to or loss of data, programs or equipment, and

unavailability or interruption of operations.

6. DISCLAIMER OF LIABILITY

EXCEPT AS EXPRESSLY SET FORTH IN THIS AGREEMENT, NEITHER RECIPIENT NOR ANY

CONTRIBUTORS SHALL HAVE ANY LIABILITY FOR ANY DIRECT, INDIRECT, INCIDENTAL,

SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING WITHOUT LIMITATION

LOST PROFITS), HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN

CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)

ARISING IN ANY WAY OUT OF THE USE OR DISTRIBUTION OF THE PROGRAM OR THE

EXERCISE OF ANY RIGHTS GRANTED HEREUNDER, EVEN IF ADVISED OF THE POSSIBILITY

OF SUCH DAMAGES.

7. GENERAL

If any provision of this Agreement is invalid or unenforceable under

applicable law, it shall not affect the validity or enforceability of the

remainder of the terms of this Agreement, and without further action by the

parties hereto, such provision shall be reformed to the minimum extent

necessary to make such provision valid and enforceable.

If Recipient institutes patent litigation against any entity (including a

cross-claim or counterclaim in a lawsuit) alleging that the Program itself

(excluding combinations of the Program with other software or hardware)

infringes such Recipient's patent(s), then such Recipient's rights granted

under Section 2(b) shall terminate as of the date such litigation is filed.

All Recipient's rights under this Agreement shall terminate if it fails to

comply with any of the material terms or conditions of this Agreement and

does not cure such failure in a reasonable period of time after becoming

aware of such noncompliance. If all Recipient's rights under this Agreement

terminate, Recipient agrees to cease use and distribution of the Program as

soon as reasonably practicable. However, Recipient's obligations under this


Agreement and any licenses granted by Recipient relating to the Program shall

continue and survive.

Everyone is permitted to copy and distribute copies of this Agreement, but in

order to avoid inconsistency the Agreement is copyrighted and may only be

modified in the following manner. The Agreement Steward reserves the right to

publish new versions (including revisions) of this Agreement from time to

time. No one other than the Agreement Steward has the right to modify this

Agreement. The Eclipse Foundation is the initial Agreement Steward. The

Eclipse Foundation may assign the responsibility to serve as the Agreement

Steward to a suitable separate entity. Each new version of the Agreement will

be given a distinguishing version number. The Program (including

Contributions) may always be distributed subject to the version of the

Agreement under which it was received. In addition, after a new version of

the Agreement is published, Contributor may elect to distribute the Program

(including its Contributions) under the new version. Except as expressly

stated in Sections 2(a) and 2(b) above, Recipient receives no rights or

licenses to the intellectual property of any Contributor under this

Agreement, whether expressly, by implication, estoppel or otherwise. All

rights in the Program not expressly granted under this Agreement are

reserved.

This Agreement is governed by the laws of the State of New York and the

intellectual property laws of the United States of America. No party to this

Agreement will bring a legal action under this Agreement more than one year

after the cause of action arose. Each party waives its rights to a jury trial

in any resulting litigation.

Eclipse Distribution License - v 1.0

Copyright (c) 2007, Eclipse Foundation, Inc. and its licensors.

All rights reserved.

Redistribution and use in source and binary forms, with or without

modification, are permitted provided that the following conditions are met:

Redistributions of source code must retain the above copyright notice, this

list of conditions and the following disclaimer.

Redistributions in binary form must reproduce the above copyright notice,

this list of conditions and the following disclaimer in the documentation

and/or other materials provided with the distribution.

Neither the name of the Eclipse Foundation, Inc. nor the names of its

contributors may be used to endorse or promote products derived from this

software without specific prior written permission.

THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND

ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED

WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE

DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR

ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES

(INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;

LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON

ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT

(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS

SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.


A

OpenSSL toolkitThis product uses OpenSSL toolkit licensed under the OpenSSL License and the Original SSLeay License.

The permission notice of the OpenSSL License and Original SSLeay License is described below. LICENSE ISSUES

==============

The OpenSSL toolkit stays under a double license, i.e. both the conditions of

the OpenSSL License and the original SSLeay license apply to the toolkit.

See below for the actual license texts.

OpenSSL License

---------------

/* ====================================================================

* Copyright (c) 1998-2018 The OpenSSL Project. All rights reserved.

*

* Redistribution and use in source and binary forms, with or without

* modification, are permitted provided that the following conditions

* are met:

*

* 1. Redistributions of source code must retain the above copyright

* notice, this list of conditions and the following disclaimer.

*

* 2. Redistributions in binary form must reproduce the above copyright

* notice, this list of conditions and the following disclaimer in

* the documentation and/or other materials provided with the

* distribution.

*

* 3. All advertising materials mentioning features or use of this

* software must display the following acknowledgment:

* "This product includes software developed by the OpenSSL Project

* for use in the OpenSSL Toolkit. (http://www.openssl.org/)"

*

* 4. The names "OpenSSL Toolkit" and "OpenSSL Project" must not be used to

* endorse or promote products derived from this software without

* prior written permission. For written permission, please contact

* [email protected].

*

* 5. Products derived from this software may not be called "OpenSSL"

* nor may "OpenSSL" appear in their names without prior written

* permission of the OpenSSL Project.

*

* 6. Redistributions of any form whatsoever must retain the following

* acknowledgment:

* "This product includes software developed by the OpenSSL Project

* for use in the OpenSSL Toolkit (http://www.openssl.org/)"

*

* THIS SOFTWARE IS PROVIDED BY THE OpenSSL PROJECT ``AS IS'' AND ANY

* EXPRESSED OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE

* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR

* PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE OpenSSL PROJECT OR

* ITS CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,

* SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT

* NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;

* LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)

* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,

* STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)

* ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED

* OF THE POSSIBILITY OF SUCH DAMAGE.

* ====================================================================

*

* This product includes cryptographic software written by Eric Young


* ([email protected]). This product includes software written by Tim

* Hudson ([email protected]).

*

*/

Original SSLeay License

-----------------------

/* Copyright (C) 1995-1998 Eric Young ([email protected])

* All rights reserved.

*

* This package is an SSL implementation written

* by Eric Young ([email protected]).

* The implementation was written so as to conform with Netscapes SSL.

*

* This library is free for commercial and non-commercial use as long as

* the following conditions are aheared to. The following conditions

* apply to all code found in this distribution, be it the RC4, RSA,

* lhash, DES, etc., code; not just the SSL code. The SSL documentation

* included with this distribution is covered by the same copyright terms

* except that the holder is Tim Hudson ([email protected]).

*

* Copyright remains Eric Young's, and as such any Copyright notices in

* the code are not to be removed.

* If this package is used in a product, Eric Young should be given attribution

* as the author of the parts of the library used.

* This can be in the form of a textual message at program startup or

* in documentation (online or textual) provided with the package.

*

* Redistribution and use in source and binary forms, with or without

* modification, are permitted provided that the following conditions

* are met:

* 1. Redistributions of source code must retain the copyright

* notice, this list of conditions and the following disclaimer.

* 2. Redistributions in binary form must reproduce the above copyright

* notice, this list of conditions and the following disclaimer in the

* documentation and/or other materials provided with the distribution.

* 3. All advertising materials mentioning features or use of this software

* must display the following acknowledgement:

* "This product includes cryptographic software written by

* Eric Young ([email protected])"

* The word 'cryptographic' can be left out if the rouines from the library

* being used are not cryptographic related :-).

* 4. If you include any Windows specific code (or a derivative thereof) from

* the apps directory (application code) you must include an acknowledgement:

* "This product includes software written by Tim Hudson ([email protected])"

*

* THIS SOFTWARE IS PROVIDED BY ERIC YOUNG ``AS IS'' AND

* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE

* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE

* ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE

* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL

* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS

* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)

* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT

* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY

* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF

* SUCH DAMAGE.

*

* The licence and distribution terms for any publically available version or

* derivative of this code cannot be changed. i.e. this code cannot simply be

* copied and put under another distribution licence

* [including the GNU Public Licence.]


A

*/


LabVIEW Runtime, LabWindows/CVI RuntimeThis product uses LabVIEW Runtime and LabWindows/CVI Runtime.

The copyright notice of LabVIEW Runtime and LabWindows/CVI Runtime is described below.Copyright 2017 National Instruments Corporation. All Rights Reserved.

APPXAppendix 6 Supported Edgecross Basic Software Version 457

A

Appendix 6 Supported Edgecross Basic Software Version

This section shows the version of Edgecross Basic Software supported by Real-time Data Analyzer.

Similar Waveform Recognition ToolThe following table shows the version of Edgecross Basic Software supported by Similar Waveform Recognition Tool.

*1 It can be checked by selecting [Help] [Version Information] in the screen shown in the following section:Page 33 Home screen

*2 For the checking method, refer to the following:Edgecross Basic Software for Windows User's Manual

Real-time Statistic Diagnosis ToolThe following table shows the version of Edgecross Basic Software supported by Real-time Statistic Diagnosis Tool.

*1 It can be checked by selecting [Help] [Version Information] in the screen shown in the following section:Page 111 Analyzer

*2 For the checking method, refer to the following:Edgecross Basic Software for Windows User's Manual

Similar Waveform Recognition Tool version*1 Edgecross Basic Software version*2

1.00A or later 1.00 or later

Real-time Statistic Diagnosis Tool version*1 Edgecross Basic Software version*2

1.00A or later 1.00 or later

458APPXAppendix 7 Added and Changed Functions

Appendix 7 Added and Changed FunctionsThis section shows the added and changed functions of Real-time Data Analyzer.

Added/changed contents Version Reference

■Similar Waveform Recognition Tool

• Data in nanoseconds can be used.

1.02C Page 62 CREATING REFERENCE WAVEFORM

LEARNING DATA

■Real-time Statistic Diagnosis Tool

• The programmable controller CPU cooperation function is

supported.

1.03D Page 394 Programmable Controller CPU Cooperation

Function

Cooperation with Real-time Flow Manager installed in another

industrial PC is supported.

1.05F Page 44 PROCEDURE TO USE Similar Waveform

Recognition Tool

Page 114 PROCEDURE TO USE Real-time Statistic

Diagnosis Tool


• A similarity diagnosis can be paused.

Page 43 Pause of Similarity Diagnosis

Page 74 Diagnosis Stop Condition Setting Screen


• Programmable controller CPU cooperation function:

A multiple regression analysis is supported.

Page 402 Multiple regression analysis


• The project file setting utility is supported.

Page 122 Project File Setting Utility


• Read data and analysis result data can be output to an

external media.



• Multiple regression analysis:

P-values can be output.



• Diagnosis information (inspection waveform and diagnosis

result) can be constantly output to a file.

1.08J Page 69 Operation Setting Screen

Page 81 When "Fixed cycle output" is selected for "Output

Settings"


• Response data:

Diagnosis results can be output.




• "Operation Settings" screen:

The similarity score threshold that corresponds to the

detection sensitivity can be displayed.



• Variables can be added to the set unit definition.

Page 148 Adding variables


• A function that edits the SPC specification values is

supported.

Page 199 Editing an SPC specification value


• If an error is detected during a simple diagnosis, detailed

information when the error occurred can be output to a file


Page 420 Diagnosis result detail information file setting

459

I

INDEX

A

Analyzer . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164Available characters . . . . . . . . . . . . . . . . . . . . 433

C

Correlation graph . . . . . . . . . . . . . . . . . . . . . . 235Creating a correlation matrix . . . . . . . . . . . . . . 222CSV file reading mode. . . . . . . . . . . . . . . . . . . 162CSV input folder . . . . . . . . . . . . . . . . . . . . . . . 129

D

Data analysis setting . . . . . . . . . . . . . . . . . . . . . 45Data diagnosis setting . . . . . . . . . . . . . . . . . . . . 50Data management setting . . . . . . . . . . . . . . . . 390Dataset . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 384Diagnosis result file . . . . . . . . . . . . . . . . . . . . . . 81Diagnosis result file display . . . . . . . . . . . . . . . . 91Displaying a diagnostic logic result . . . . . . . . . . 382Displaying a result of simple diagnosis . . . . . . . 374Displaying data after running . . . . . . . . . . . . . . 365Displaying read data . . . . . . . . . . . . . . . . . . . . 165Displaying read data frequency . . . . . . . . . . . . 183Displaying read data statistics . . . . . . . . . . . . . 180

E

Editing Logics . . . . . . . . . . . . . . . . . . . . . . . . . 255Equipment data . . . . . . . . . . . . . . . . . . . . . . . 225Extended section . . . . . . . . . . . . . . . . . . . . . . 155

F

Function list . . . . . . . . . . . . . . . . . . . . . . . . . . . 20Functional specifications . . . . . . . . . . . . . . . . . . 14

G

Guard band diagnosis . . . . . . . . . . . . . . . . . . . 215

I

Incorrect variable . . . . . . . . . . . . . . . . . . . . . . 227

L

License Manager . . . . . . . . . . . . . . . . . . . . . . . 24Logic setting . . . . . . . . . . . . . . . . . . . . . . . . . . 393

M

Maharanobis - Taguchi method (MT). . . . . . . . . 250Management . . . . . . . . . . . . . . . . . . . . . . . . . 384Measurement data . . . . . . . . . . . . . . . . . . . . . 228Monitor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 416MT distance calculation . . . . . . . . . . . . . . . . . . 252Multiple Regression Analysis (LMR) . . . . . . . . . 243Multivariate analysis (single central value format) 207

O

Offline analysis. . . . . . . . . . . . . . . . . . . . . . . . . .12Operational specifications . . . . . . . . . . . . . . . . . .19

P

Partial matrix . . . . . . . . . . . . . . . . . . . . . . . . . .231Primitive . . . . . . . . . . . . . . . . . . . . . . . . . 255,261Procedure before operation . . . . . . . . . . . . . . . . .23Programmable controller CPU cooperation function394

R

Reader . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .125Reader equipment data folder . . . . . . . . . . . . . .129Real-time diagnosis . . . . . . . . . . . . . . . . . . . . . .13Recurrence calculation . . . . . . . . . . . . . . . . . . .247Reference waveform learning data . . . . . . . . . . . .36Response data file . . . . . . . . . . . . . . . . . . . . . .108RSD dataset folder . . . . . . . . . . . . . . . . . . . . . .129

S

Section level . . . . . . . . . . . . . . . . . . . . . . . . . .267Setting a logic variable . . . . . . . . . . . . . . . . . . .356Similar waveform recognition monitor . . . . . . . . . .90Similarity score. . . . . . . . . . . . . . . . . . . . . . . . . .41Similarity score threshold list . . . . . . . . . . . . . . . .42SPC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .189SPC rule . . . . . . . . . . . . . . . . . . . . . . . . . . . . .192SPC specification value setting file . . . . . . . . . . .198

T

Text variable value . . . . . . . . . . . . . . . . . . . . . .139Troubleshooting by symptom . . . . . . . . . . . . 98,422

U

Unit definition . . . . . . . . . . . . . . . . . . . . . . . . . .131Unit section . . . . . . . . . . . . . . . . . . . . . . . . . . .131

W

Waveform display setting. . . . . . . . . . . . . . . . . .391Whole matrix . . . . . . . . . . . . . . . . . . . . . . . . . .225

460

MEMO

461

REVISIONS*The manual number is given on the bottom left of the back cover.

Japanese manual number: SH-081870-L

2018 MITSUBISHI ELECTRIC CORPORATION

Revision date *Manual number Description

May 2018 SH(NA)-081873ENG-A First edition

May 2018 SH(NA)-081873ENG-B Partial correction

September 2018 SH(NA)-081873ENG-C ■Added or modified parts

Section 16, Section 19.4, Appendix 2, Appendix 3, Appendix 5

March 2019 SH(NA)-081873ENG-D ■Added or modified parts

Section 2.1, Section 3.2, Section 4.3, Section 9.1, Section 11.1, Section 13.1, Section 13.2, Chapter

17, Section 19.1, Section 19.2, Section 19.3, Section 19.4, Section 19.5, Section 19.6, Chapter 20,

Section 20.1, Section 20.2, Section 20.3, Section 20.4, Section 20.5, Section 20.6, Section 20.7,

Section 20.8, Section 20.9, Section 20.10, Section 20.11, Section 20.12, Section 20.13, Section

20.14, Section 20.15, Section 20.16, Section 20.17, Section 20.18, Section 20.19, Section 21.1,

Section 21.2, Section 21.3, Appendix 2, Appendix 4, Appendix 6, Appendix 7

March 2019 SH(NA)-081873ENG-E ■Added or modified parts

Section 19.4, Section 19.5, Section 20.6

April 2019 SH(NA)-081873ENG-F Partial correction

June 2019 SH(NA)-081873ENG-G ■Added or modified parts

Section 16, Appendix 4

December 2019 SH(NA)-081873ENG-H ■Added or modified parts

Section 2.1, Section 3.1, Chapter 4, Chapter 7, Section 7.4, Chapter 8, Section 8.1, Section 8.2,

Section 10.1, Section 11.1, Section 11.3, Chapter 12, Section 12.1, Section 14.1, Section 14.2,

Chapter 16, Chapter 18, Section 18.1, Section 18.2, Section 18.3, Section 18.4, Section 19.1,

Section 19.3, Section 19.4, Section 20.1, Section 20.5, Section 20.6, Section 20.7, Section 20.8,

Section 20.19, Chapter 22, Appendix 1, Appendix 2, Appendix 4, Appendix 7

July 2020 SH(NA)-081873ENG-I ■Added or modified parts

Section 4.1, Section 14.1

March 2021 SH(NA)-081873ENG-J ■Added or modified parts

Section 2.1, Section 5.1, Section 8.2, Section 8.4, Chapter 11, Section 11.1, Chapter 12, Section

12.1, Chapter 13, Section 13.1, Chapter 14, Section 14.1, Section 14.2, Section 14.3, Section 14.4,

Section 15.2, Chapter 17, Section 19.2, Section 20.2, Section 20.3, Section 20.5, Section 21.4,

Section 21.10, Chapter 22, Section 22.2, Section 22.3, Appendix 1, Appendix 2

June 2021 SH(NA)-081873ENG-K ■Added or modified parts

Section 20.4, Section 20.5

This manual confers no industrial property rights of any other kind, nor does it confer any patent licenses. Mitsubishi Electric Corporation cannot be held

responsible for any problems involving industrial property rights which may occur as a result of using the contents noted in this manual.

462

TRADEMARKSIntel is either registered trademarks or trademarks of Intel Corporation in the United States and/or other countries.

Microsoft, Windows, and Excel are either registered trademarks or trademarks of Microsoft Corporation in the United States

and/or other countries.

Sentinel is a registered trademark of SafeNet, Inc.

The company names, system names and product names mentioned in this manual are either registered trademarks or

trademarks of their respective companies.

In some cases, trademark symbols such as '' or '' are not specified in this manual.

SH(NA)-081873ENG-K(2106)

MODEL: SW1DND-RDA-U-E

Specifications subject to change without notice.

When exported from Japan, this manual does not require application to theMinistry of Economy, Trade and Industry for service transaction permission.

HEAD OFFICE : TOKYO BUILDING, 2-7-3 MARUNOUCHI, CHIYODA-KU, TOKYO 100-8310, JAPANNAGOYA WORKS : 1-14 , YADA-MINAMI 5-CHOME , HIGASHI-KU, NAGOYA , JAPAN

Real-time Data Analyzer User's Manual

Documents

Transcript of Real-time Data Analyzer User's Manual